Search

Junction termination extension (JTE) is a mature edge termination structure in Si and SiC power devices, but remains a technical challenge in GaN due to lack of mature selective area doping technique. In this work, GaN vertical p-n diodes with Hydrogen-Modulated Step Graded JTE (HMSG-JTE) are realized by gradually enlarging photolithography windows for hydrogen plasma treatment and controlled thermal diffusion of hydrogen in p-GaN. HMSG-JTE leads to a gradually decreasing hole distribution outwards from the main junction, which laterally spreads the electric field and mitigates the electric field peak under reverse bias. Consequently, the breakdown voltage of vertical GaN p-n diode is successfully boosted from 661 V (with hydrogen-plasma-based edge termination) to 1489 V with HMSG-JTE.

Oral lichen planus (OLP) is a chronic inflammatory oral mucosal disease with unclear etiology. Autoimmune thyroid diseases (AITD) is a type of autoimmune disease characterized by increased thyroid-specific antibodies. In recent years, more and more studies have found that the incidence of AITD is increased in OLP patients. The occurrence and development of OLP and AITD may be related to the expression of thyroid autoantigen in oral keratinocytes, the imbalance of thyroid hormone （Th）1/Th2 and Th17/Treg cell subsets, the abnormal quantity and function of follicular helper T cells and chemokines and the specific killing ability of CD8 T cells to target cells. In this article, the possible immune mechanisms involved in the coexistence of OLP and AITD are reviewed to provide insights for the diagnosis, treatment and prevention of these two diseases from the perspective of immunology.

AlGaN/GaN high-electron-mobility transistor (HEMT) based pH sensors have the advantages of fast response and high stability, and can be used in harsh environments. This paper presents a pH sensor based on a planar dual-gate AlGaN/GaN HEMT cascode amplifier, which can increase the pH sensitivity for about 45 times from 45 mV/pH to 2.06 V/pH with linearity of 1.27%. The results indicate that it is possible to improve the pH sensitivity at the origin of detecting instead of the subsequent complex amplifier circuits. The proposed device has also shown the capacity to adjust the sensor’s sensitivity by changing the gate voltage and the resistance values for various pH ranges of different measuring requirements.

An L-shaped two-dimensional acoustic particle velocity sensor (APVS), which has integrated two best performing one-dimensional APVS on a single chip is presented. This structure keeps the two sensitive axes exactly orthogonal to each other, avoiding the alignment error. There are no silicon substrate blocking in both sensitive directions, so the flow directions nearby the wires have little differences. As a result, the two axes have equal sensitivity, thus requiring no correction. The measured velocity sensitivity at 1 kHz is 2.53 dB V/Pa, with relative direction sensitivity of 28 dB. Since the total size of the chip is only 4.4 × 4.4 mm, it can detect the 2-D particle velocity at one point.

In this letter, a distributed network model describing the effects of the border traps and distributed channel resistance on the impedance frequency dispersion of lateral MOS devices is proposed. The proposed model is verified using a gate recessed, normally-off Al2O3/GaN MOSFET structure operating as a MOS diode. The measured frequency-dependent capacitance and conductance curves of the MOS diode over a wide frequency range are found to be in good agreement with the proposed model. According to the intrinsic property of border traps to the ac signal, the proposed model is further modified to get the spatial distribution of border traps. The new insight derived from the impedance dispersion characteristics of lateral MOS devices is critical for quantitative analysis of the quality of III-V lateral MOS structures.

Oral squamous cell carcinoma (OSCC) is one of the most common cancers worldwide, and with 354 864 new cases each year. Cancer metastasis, recurrence, and drug resistance are the main causes to cripples and deaths of OSCC patients. As potent growth factors, fibroblast growth factors (FGFs) are frequently susceptible to being hijacked by cancer cells. In this study, we show that FGF8 is upregulated in OSCC tissues and high FGF8 expression is related with a set of clinicopathologic parameters, including age, drinking, and survival time. FGF8 treatment enhances the invasive capability of OSCC cells. Lentivirus-based FGF8 expression promotes OSCC metastasis in a mouse lung metastasis model. Further, mechanistic study demonstrates that FGF8 induces epithelial–mesenchymal transition (EMT) in OSCC cells. These results highlight a pro-metastatic function of FGF8, and underscore the role of FGF8 in OSCC development.

Electric field crowding at the Schottky edge and Schottky barrier lowering induced leakage current are two major problems that limit the performance of GaN-on-GaN vertical SBD. In this paper, we try to solve these issues. On one hand, a planar edge termination method based on high-energy fluorine ion implantation is explored for high voltage vertical GaN SBD. The space charge induced field modulation effect (SCIFM) is discovered in the ion implanted region. The electrons injected in the implanted region could effectively reduce the surface electric field, alleviating the field crowding at the metal edge. The SBD terminated with high-energy fluorine implanted region presents a breakdown voltage of 1300 V. Utilizing the depletion effect of space charges, the leaky path through the implanted region is partially blocked by the high-energy fluorine ion implanted guard rings. The reverse leakage current of SBD with guard rings is reduced by nearly 3 orders at the reverse bias of 300 V without the deterioration of forward characteristics. On the other hand, the fluorine ion implantation based quasi-JBS structure is investigated to shield the electric field seen by the Schottky contact and reduce the Schottky barrier lowering induced leakage current. The turn-on voltage modulation effect and improved breakdown voltage are observed in the fabricated quasi-JBS diodes, confirming the validity of the SCIFM mechanism in the ion implanted region.

The great advances in silicon photonic sensing technology have made it an attractive platform for a wide of sensing applications. However, most silicon photonic sensing platforms suffer from high susceptibility to the temperature fluctuation of operating environment. Additional complex and costly chemical signal enhancement strategies are usually required to improve the signal-to-noise ratio (SNR). Here, a biotoxoid photonic sensor that is resistant to temperature fluctuation have been demonstrated. This novel sensor consists of a ring resonator coupled to a Mach-Zehnder interferometer (MZI) readout unit. Instead of using costly wavelength interrogation, our photonic sensor directly measures the light intensity ratio between the two output ports of MZI. The temperature dependence (TD) controlling section of the MZI is used to eliminate the adverse effects of ambient temperature fluctuation. The simulation and experimental results show a linear relationship between the interrogation function and the concentration of analyte under operation conditions. The thermal drift of the proposed sensor is just 0.18%, which is a reduction of 567 folds for chemical sensing and 28 folds for immuno-biosensing compared to conventional single ring resonator. The SNR increases from 6.85 dB to 19.88 dB within a 2°C temperature variation. The high SNR optical sensor promises great potential for amplification-free of detection of nucleic acids and other biomarkers. Ministry of Education (MOE) National Research Foundation (NRF) Accepted version

A series of studies are dedicated to research the clinical outcomes of oral leukoplakia (OLK) treated with carbon dioxide laser (CO2 laser); however, the results vary from studies especially related to recurrence and malignant transformation. Hence, we performed this meta-analysis to precisely evaluate the malignant transformation of OLK dealt with CO2 laser and investigate the association between its malignant transformation and kinds of related risk factors, such as gender, clinical classification, long duration of leukoplakia, and degree of epithelial dysplasia and lesion regions. We performed a systematic search of the Cochrane Library, EMBASE, Pubmed, Web of Science, and SCOPUS. Single-arm rate of the overall risk of malignant transformation in OLK treated with CO2 laser was calculated using the Der-Simonian Liard method. We applied subgroup analysis to compare the risk of malignant transformation according to the degree of epithelial dysplasia, clinical type, and region of OLK. Moreover, a pooled odds ratio (OR) is calculated, along with its 95% confidence interval (CI), to compare the risk of malignant transformation according to patients' gender, tobacco, and alcohol consumption. We used the meta package of R software for quantitative data synthesis and analysis. The rate of malignant transformation of OLK treated with carbon dioxide laser ranged from 0 to 15.38% in included studies. The overall rate of malignant transformation of OLK treated with CO2 laser is 4.50% under the random effect model [95% CI 0.0305-0.0659]. A systematic review of observational studies of OLK reported that the estimated overall (mean) malignant transformation rate was 3.5%, with a wide range between 0.13 and 34.0%. Interestingly, our result revealed that it was the male, homogeneous type, no tobacco consumption, and without alcohol-use who had a higher tendency of malignancy after laser surgery. However, this result lack statistically significant data. Generally speaking, whether oral leukoplakia patients underwent laser surgical treatment or not, it may have little effect on malignant transformation. In addition, we strongly advise that it had better not to perform CO2 laser intervention on OLK patients with the following clinical characteristics: homogeneous type, male, no tobacco consumption, and without alcohol-use. Evidence is still lacking in terms of relationship between malignant transformation and risk factors among OLK patients managed with CO2 laser. Thus, these associations should be further investigated.

In this paper, we report the device performance of a high-voltage enhancement-mode (E-mode) GaN MOSHEMT on silicon substrate. Normally off operation is realized by a self-terminated precision gate recess process on an optimized high-electron mobility transistor structure. The GaN MOSHEMT is fully pinched off at zero gate bias, suggesting a “true” normally off operation. The threshold voltage is 0.4 V with a drain current density of 1 $\mu \text{A}$ /mm as the criteria. The device with 15- $\mu \text{m}$ gate-drain distance and 100- $\mu \text{m}$ gate width exhibits a maximum drain current of 356 mA/mm at 8-V gate bias. The on/off current ratio of the device is larger than 1010 with a subthreshold slope of 80 mV/dec. The gate leakage current is below 10−7 mA/mm up to 9-V gate bias. The off-state breakdown voltage (BV) is as high as 1528 V (880 V) measured with floating (grounded) silicon substrate at a drain leakage current criterion of 5 $\mu \text{A}$ /mm. The specific on-resistance ( ${R}_{\text {ON,SP}}$ ) of the device is 2.79 $\text{m}\Omega \cdot$ cm2, and the power figure of merit (BV2/ ${R}_{ \text {ON,SP}}$ ) is 277 MW/cm2. High-voltage pulsed I–V measurement indicates that the dynamic on-resistance is only 1.6 times the static one with a pulsewidth of 10 $\mu \text{s}$ at 400-V off-state quiescent drain bias. The high performance of the normally off GaN MOSHEMT is supposed to benefit from the high quality low pressure chemical vapor deposition Si3N4 passivation layer and the advanced E-mode device fabrication process.

Antibiotic resistance genes (ARGs) are ubiquitous in the environment, with previous studies mainly focusing on the terrestrial ecosystem, which is prone to higher antibiotic application. However, the characteristics, distribution pattern, and driving factors of soil ARGs at the macro scale are still unclear. In this study, the soil ARGs, antibiotics, mobile genetic elements (MGEs), soil properties, toxic metals, polycyclic aromatic hydrocarbons (PAHs), and bacterial community in the Taipu River Basin were analyzed to investigate the distribution and dissemination of ARGs at a watershed scale. The results revealed that ARGs were widespread in the soils along the Taipu River, and that ARG profiles varied greatly with different types of land use, but showed regional similarities. The characteristics were mainly determined by antibiotic input and the ARG transmission mediated by MGEs. The order of the contribution of environmental factors to ARG distribution was toxic metals &gt, PAHs &gt, soil properties. Toxic metal pollution was coupled with ARGs through MGE mediation, while PAHs and soil properties were most likely to affect the ARG distribution by shifting the bacterial community. The microbial–ecological relationship changed significantly with the enrichment of ARGs, and its impact may extend to the watershed scale. Transposon IS1247 can be used as an indicator of the ARGs impact on the microbial ecological relationship in the soils of the Taipu River Basin.

For devices with a 15 micron anode-to-cathode distance, nearly 1.5 times increase in the blocking (breakdown) voltage (from 692 to 1030 V) has been achieved by replacing the alloyed Ohmic contact at the anode electrode of the conventional MOS gated hybrid-anode lateral field-effect rectifier (CMLFER) with a low barrier Schottky contact. The new Schottky-MOS hybrid-anode lateral field-effect rectifier is found to offer comparable low onset voltage ( ${V}_{{\textbf {ON}}}$ of 0.68±0.13 versus 0.65±0.11 V for CMLFER) independent of the anode-to- cathode distance. The immunity of the punch through caused by drain induced barrier lowering effect is obtained through the low barrier Schottky contact in anode, which is believed to be responsible for the reduction in the leakage current, and the improvement of rectifier breakdown voltage.

In this letter, a gate recessed normally-off GaN metal–oxide–semiconductor high-electron-mobility transistor on silicon substrate is fabricated using AlN/Si3N4 as the passivation layer. The thin AlN layer serves the dual role of protecting the gate channel region from direct plasma bombardment during the RIE Si3N4 removal and passivating the surface states in the access region. As a result, the effective carrier mobility in the normally-off channel is found to improve from the 568 cm2/ $\text {V}\cdot \text {s}$ in conventional Si3N4 passivation process to a high value of 1154 cm2/ $\text {V}\cdot \text {s}$ . A saturated output current density of 603 mA/mm and an ON-resistance of $5.3~\Omega \cdot \text {mm}$ was obtained for devices with $\text{L}_{\text {G}}/\text{L}_{\text {GS}}/\text{L}_{\text {GD}}/\text{W}_{\text {G}} = 1.5$ /1.5/3/ $20~\mu \text{m}$ . Meanwhile, the degradation of dynamic ON-resistance is significantly suppressed due to the effective passivation of surface states by the AlN layer grown by plasma-enhanced atomic layer deposition.

// Ruixue Ai 1 , Yan Tao 1 , Yilong Hao 1 , Lu Jiang 1 , Hongxia Dan 1 , Ning Ji 1 , Xin Zeng 1 , Yu Zhou 1 and Qianming Chen 1 1 State Key Laboratory of Oral Diseases, National Clinical Research Center for Oral Diseases, Department of Oral Medicine of West China Hospital of Stomatology, Sichuan University, Chengdu, Sichuan, China Correspondence to: Yu Zhou, email: zhouyu19830306@sina.com Qianming Chen, email: qmchen@scu.edu.cn Keywords: immunity, premalignant condition, microenvironment, oral potentially malignant disorder, therapeutics Received: May 24, 2017 Accepted: August 04, 2017 Published: August 17, 2017 ABSTRACT Oral potentially malignant disorders (OPMD) develop in a complex tissue microenvironment where they grow sustainably, acquiring oral squamous cell carcinoma (OSCC) characteristics. The malignant tumor depends on interactions with the surrounding microenvironment to achieve loco-regional invasion and distant metastases. Unlike abnormal cells, the multiple cell types in the tissue microenvironment are relatively stable at the genomic level and, thus, become therapeutic targets with lower risk of resistance, decreasing the risk of OPMD acquiring cancer characteristics and carcinoma recurrence. However, deciding how to disrupt the OPMD and OSCC microenvironments is itself a daunting challenge, since their microenvironments present opposite capacities, resulting in diverse consequences. Furthermore, recent studies revealed that tumor-associated immune cells also participate in the process of differentiation from OPMD to OSCC, suggesting that reeducating stromal cells may be a new strategy to prevent OPMD from acquiring OSCC characteristics and to treat OSCC. In this review, we discuss the characteristics of the microenvironment of OPMD and OSCC as well as new therapeutic strategies.

Time-dependent OFF-state leakage behavior of AlGaN/GaN MISHEMTs on silicon substrate is investigated and a novel degradation mechanism is proposed in this paper. Under constant high voltage OFF-state stress, drain leakage current gradually increases with stress time and the behavior is gate bias and temperature-dependent. Consecutive OFF-state breakdown measurement with drain injection technique indicates that the negative shift of threshold voltage ( $V_{\mathrm {th}}$ ) is responsible for the increase of drain leakage current during stress measurement. It is proposed that the negative shift of $V_{\mathrm {th}}$ is mainly induced by the ionization of uncompensated donor like deep levels near the channel, which are most likely to be located in the 300-nm-thick unintentionally doped GaN layer above the carbon doped buffer layer.

In this letter, based on a new growth strategy, we demonstrate a high-performance quasi-vertical GaN Schottky barrier diode (SBD) fabricated on the hetero-epitaxial layer on silicon, with the highest current on/off ratio of 1010, the lowest specific on-resistance of 0.95 mΩ·cm2, and the lowest ideality factor of 1.23 among the GaN quasi-vertical GaN SBD reported until now. It reveals the great potential of the GaN vertical structure device on Si thanks to the novel growth strategy, which is capable of realizing low density of threading dislocation and thick epitaxial layer at the same time.

Robust gate dielectric and dielectric/semiconductor are highly desired for GaN power MOS devices, while the trapping effect of the border traps close to the MOS interface could lead to reliability issues in those devices. In this paper, a complete model for border traps in lateral GaN MOS diode is proposed based on a distributed network of the border traps and channel resistance. We show that excluding the effect of channel resistance is critical for accurate calculating the distribution of border traps in lateral devices with low channel mobility, such as GaN and SiC MOS devices. The proposed model agrees well with the measured frequency dependent capacitance and conductance curves of Al 2 O 3 /GaN MOS diode in a gate recessed normally-off GaN power MOSFET. The new insight derived from the impedance dispersion characteristics of lateral MOS devices is critical for quantitative analysis of the quality of III-V lateral MOS structures.

This paper proposed a novel MEMS 3-axis thermal accelerometer for the first time. Two five-wire structures perpendicular to each other are fabricated using planar stacking method. Compared with other 3-axis thermal accelerometers, this accelerometer has simpler fabrication process and signal processing circuit, and the sensitive directions are orthogonal with almost the same sensitivities. For one of the five-wire structures, the sensitivities of the two orthogonal directions are 16.1mV/g and 18.4mV/g respectively. This accelerometer shows good linearity for a range about 0-6g acceleration at 20Hz.

This paper presents an original design of a hybridized metal/polydimethylsiloxane (PDMS) sponge for multi-directional pressure energy harvesting. This kind of wired-sponge triboelectric nanogenerator (wired-sponge TENG) has the characterizations of stable structure, simple fabrication, low expense and high performance. The voltage signal output has a maximum peak-to-peak value of 45 V and the current signal output has a maximum peak-to-peak value of 400 nA, when the wired-sponge TENG is compressed to 50% of strain under the frequency of 1 Hz. And the maximum average output power of TENG reaches 0.25 μW under the external load of 100KΩ at the same experimental circumstance.

Carbon doping in the buffer of AlGaN/GaN high-electron-mobility transistors (HEMTs) leads to the notorious current collapse phenomenon. In this paper, an HEMT structure with a source-connected p-GaN (SCPG) embedded in the carbon-doped semi-insulating buffer is proposed to suppress the buffer-induced current collapse effect. Two-dimensional transient simulation was carried out to show the successful suppression of buffer-induced current collapse in the SCPG-HEMTs compared with conventional HEMTs. The mechanism of suppressing dynamic on-resistance degradation by ejecting holes from the SCPG into the high resistive buffer layer after off-state stress is illustrated based on energy band diagrams. This paper contributes an innovative device structure to potentially solve the buffer-induced degradation of the dynamic on-resistance in GaN power devices.

In this letter, a plasma-free etch stop structure is developed for GaN HEMT toward enhancement-mode operation. The self-terminated precision gate recess is realized by inserting a thin AlN/GaN bilayer in the AlGaN barrier layer. The gate recess is stopped automatically at the GaN insertion layer after high-temperature oxidation and wet etch, leaving a thin AlGaN barrier to maintain a quantum well channel that is normally pinched off. With addition of an Al2O3 gate dielectric, quasi normally OFF GaN MOSHEMTs have been fabricated with high threshold uniformity and low ON-resistance comparable with the normally ON devices on the same wafer. A high channel mobility of 1400 cm $^{2}/\textrm {V}\cdot \textrm {s}$ was obtained due to the preservation of the high electron mobility in the quantum-well channel under the gate.

As an environmentally-friendly aquaculture and planting system, aquaponics has attracted attention in various fields, such as fisheries, agriculture, and ecology. The existing review qualitatively described the development and challenges of aquaponics but lacked data support. This study selected 513 related documents (2000–2019) in the Web of Science database (WOS) to mine and quantitatively analyze its text data. The keyword co-occurrence network shows that the current aquaponics research mainly focuses on the system components, wastewater treatment, nutrient management, and system production. Research areas reflect obvious regional characteristics. China, the United States and Europe are dedicated to the application of new technologies, the optimization of system production, and the exploration of multiple roles. At present, the aquaponics development is facing many pressures from management and market. Future research requires more in-depth research in the system construction, nutrient management, and microbial community structure to provide a theoretical basis. Moreover, the identity construction within the conceptual framework of green infrastructure is a research direction worth exploring to solve low social recognition for aquaponics.

Background:Previous studies indicate that cancers in DM/PM patients are associated with increased mortality. Hence, identifying predictors of malignancy in PM and DM is crucial. However, few large series studies have reported prognostic and predictive factors of malignancy in patients with PM and DM. Moreover, in recent years, several published studies also allow us to better understand the clinical characteristics of malignancy in PM and in DM.Objectives:To analyze the mortality and identify the major independent risk factors for death in patients with dermatomyositis/polymyositis (DM/PM) complicated with malignant tumor.Methods:The clinical data of all patients with DM/PM in Peking University First Hospital from January 2007 to Jan 2019 were retrospectively reviewed. All patients were followed up to confirm whether they had malignant tumors. According to the statistics of the National Bureau of Statistics of China, the standard mortality (SMR) and life lost years (YLL) of patients with DM/PM were combined with malignant tumors. The Kaplan-Meier method was used to analyze the 10-year survival of DM/PM patients with malignant tumors. Cox multivariate regression was used to predict independent risk factors for DM/PM patients with malignant tumors.Results:A total of 334 patients with dermatomyositis and 69 patients with polymyositis were enrolled in the study. The mean age of onset was 50.5 ± 14.8 years and 48.9 ± 16.1 years, with a median follow-up of 40.6 (11.6-77.6) months. Among them, 320 patients were successfully followed up, including 69 patients with death, 46 DM/PM with malignant tumors (38 with dermatomyositis and 8 with polymyositis). The average age of onset of DM/PM patients with malignant tumors was 55.4 ± 15.1 years and 59.5 ± 4.7 years, respectively, of which 17 died. The age-sex adjusted SMR of DM/PM patients without malignant tumors was 9.0 (95% CI 6.8-11.2). The age-sex adjusted SMR of DM/PM with malignant tumors was 12.3 (95% CI 9.0-14.7). The life loss of male patients with dermatomyositis complicated with malignant tumors was 30.1 years, and that of females was 38.6 years; the life loss of male patients with polymyositis was 27.6 years, and that of females was 22.1 years. A 10-year survival analysis showed that DM/PM patients with malignant tumor had significantly worse prognosis than patients without malignancy (p=0.001 Log-rank). The 1-, 5-, and 10-year survival rates of DM/PM patients who did not have malignant tumors were 87.9%, 81.9%, and 78.4%, respectively. DM/PM Patients with malignant tumors 1, 5, and 10 years The survival rates were 73.3%, 56.0%, and 45.7%, respectively. The independent risk factors for death in DM/PM patients with malignant tumors were advanced age (HR=1.11 95% CI 1.02-1.20, p=0.014) and infection (HR=17.07 95% CI 1.66-175.75, p= 0.017).Conclusion:Malignant tumor is a common in patients with DM/PM, and the mortality of DM/PM patients with malignant tumors is high. The independent predictors of mortality for PM/DM patients with malignant tumors were age at disease onset and infection.Acknowledgments:We thank our patients with PM/DM participating in this study.Disclosure of Interests:None declared

Oral lichen planus (OLP) is a chronic mucosal inflammatory disease. The World Health Organization has described it as a potentially malignant condition. The pathogenesis of OLP remains to be fully elucidated, but extensive evidence suggests that immunologic and inflammatory factors have important roles. MicroRNAs (miRs), which are small non-coding RNAs, have been reported to be involved in OLP. In particular, miR-155 is significantly upregulated in patients with OLP. miR-155 has numerous functions and is closely linked to inflammation and immune system regulation. However, in-depth studies of the mechanisms via which miR-155 is involved in OLP are currently insufficient. Considering the close association between miR-155 and immune regulation as well as the importance of immune factors in OLP, the role of miR-155 in the immune system was herein summarized with a focus on OLP. The present review provides a basis for further study of the molecular mechanisms underlying the development and progression of OLP.

The temperature dependence of current collapse (CC) in AlGaN/GaN high-electron mobility transistors on silicon substrate is studied in this paper. Devices without and with Si3N4 passivation are used to investigate the behavior of surface- and buffer-induced CC, respectively. It is found that the degree of surface-induced CC in unpassivated devices has a weak temperature dependence, which is induced by the cancelling out between enhanced carrier injection based on surface hopping and enhanced emission when the temperature is increased. On the other hand, the degree of buffer-induced CC in the Si3N4 passivated devices is reduced at higher temperature since the energy of hot electrons is reduced due to the phonon scattering and the trapping of hot electrons in the buffer is mitigated. Temperature-dependent transient measurement is also carried out to investigate the recovery process for these two type of CC. Two types of trap levels are identified in the unpassivated and Si3N4 passivated devices, respectively. The trap level $E1$ with an activation energy of 0.08 eV is supposed to be related to the surface trapping, while $E2$ with an activation energy of 0.22 eV is located in the buffer layer.

An original design of a self-adaptive nonlinear system (SANS) for shock protection of MEMS is presented here for the first time. The internally integrated passive silicon structure, with increasing amplitudes of shock, realizes nonlinear growth in shock resistance and reveals positive selectivity and compatibility, through three energy dissipation modes. This approach enables a generic batch fabrication requiring no additional processes or excessive area expansion (290μm×320μm for an 8cm×8cm device in this work). The SANS has been verified to provide enhanced shock robustness over conventional shock protective structures, namely the hard stop (∼2 times) and the flexible spring stop (∼1.5 times).

This work reports a kilovolt and low current collapse normally-off GaN MOSHEMT on silicon substrate. The device with a drift length of 3 μm features a threshold voltage of 1.7 V and an output current of 430 mA/mm at 8 V gate bias. The off-state breakdown voltage (BV) is as high as 1021 V (800 V) defined at a drain leakage criterion of 10 μA/mm with floating (grounded) substrate. The corresponding breakdown electric field is 3.4 MV/cm and the Baliga's figure-of-merit (BFOM) is 1.6 GW/cm2. A small degradation of the dynamic on-resistance (R on, d ) about 30% is observed with a short pulse width of 500 ns and a quiescent drain bias of 60 V. The record value is supposed to benefit from the intrinsic step-graded field plate, high quality LPCVD SI3N4 passivation and material optimization of 4.5 μm thick epitaxial layer.

In this paper, we present the theoretical analysis and calculation results for characterization of a graphene diaphragm applied for capacitive ultra-low pressure sensor structure, where a few layers of graphene diaphragm with radius of 8 µm and total thickness of 2 nm, the graphene diaphragm is suspended over a circular cavity. In this paper, the graphene film was fabricated by an atmospheric-pressure chemical vapor deposition approach to directly form large-area uniform graphene film on silicon dioxide layer. Calculation results show that this capacitive pressure sensor is able to provide 164aF/Pa pressure sensitivity in an ultralow pressure range from 1Pa to 5Pa.

This paper shows measurement of very low speed air flow with a new sensor. This micro-machined sensor contains two monolithically integrated orthogonal cells making it prospective to realize three-dimensional (3D) detection of the direction and velocity of air flow at one point. Each cell (vent) has a five-wire configuration, formed by five resistive beams of platinum (with pitch of 354 µm), sensitive to two direction. It was measured from 16.7 mm/s to 167 mm/s, showing good repeatability and demonstrating its directionality. Velocity distribution was simulated, and results corresponded to the theoretical lower limit of velocity of 50 mm/s due to natural convection, the induced angle dependent offset voltage of which can be taken into account to make a more accurate computation of lower velocity.

In this paper, we report the device performance of a high-voltage normally off ${\rm Al}_{2}{\rm O}_{3}/{\rm GaN}$ MOSFET on the Si substrate. Normally off operation is obtained by multiple cycles of ${\rm O}_{2}$ plasma oxidation and wet oxide-removal gate recess process. The recessed normally off GaN MOSFET with 3 $\mu{\rm m}$ gate-drain distance exhibits a maximum drain current of 585 mA/mm at 9 V gate bias. The threshold voltage of the MOSFET is 2.8 V with a standard derivation of 0.2 V on the sample with an area of 2 $\,\times\,$ 2 ${\rm cm}^{2}$ . The gate leakage current is below $10^{-6}~{\rm mA}/{\rm mm}$ during the whole gate swing up to 9 V and the ${\rm I}_{{\scriptstyle{\rm ON}}}/{\rm I}_{{\scriptstyle {\rm OFF}}}$ ratio is larger than $10^{9}$ , indicating the good quality of ${\rm Al}_{2}{\rm O}_{3}$ gate insulator. The MOSFET with 10 $\mu{\rm m}$ gate-drain distance shows a three terminal OFF-state breakdown voltage (BV) of 967 V at zero gate-source bias with a drain leakage current criterion of 1 $\mu{\rm A}/{\rm mm}$ . The specific ON-resistance $(R_{{{\scriptstyle {\rm ON}}},{\rm SP}})$ of the device is 1.6 ${\rm m}\Omega\,{\cdot}\,{\rm cm}^{2}$ and the power figure of merit $({\rm BV}^{2}/R_{{{\scriptstyle {\rm ON}}},{\rm SP}})$ is 584 ${\rm MW}/{\rm cm}^{2}$ .

This paper reports a digital controller based on a three dimensional adaptive filter demodulator (AFD) for micromachined vibratory gyroscopes with the goal of eliminating common-mode noise and reducing hardware resources. The least mean square (LMS) adaptive filter, which has advantages of fast convergence speed, lower noise and fewer occupied hardware resources, is adopted to demodulate the vibration velocity of the gyroscope and detect its phase shift. A three dimensional AFD is proposed to eliminate the common-mode noise and quadrature coupling induced by the initial capacitance mismatch. Simulation and experimental results have verified the effectiveness of this method. The measurement results of the digital controlled gyroscope show a zero bias drift of 24.6 °/h and a nonlinearity of 0.1% with the measurement range of ±200°/s.

The equivalent RLC circuit for the resonant accelerometer isdescribed in detail, which shows that equivalent circuit containstwo variable capacitances and two variable resistances due to theacceleration and amplitude. Using the electromechanical couplingfactor presented here and the mode function of the transversedeflection vibration of the double clamped beam, the equivalent circuit parameters related to the size of the accelerometer and thedrive voltage are derived. In short, the mechanical and electrical parameters of the resonant accelerometer can be calculated andanalyzed accurately by the formulae derived in this paper.

A differential capacitive accelerometer with simple process is designed, simulated, and fabricated. To achieve a precision structure dimension with fewer processing steps, the silicon device layer transfer technology is being used to built a sandwich accelerometer based on a silicon-on-insulator (SOI) wafer, which was assembled by glass-si-glass multilayer anodic bonding. Deep reactive ion etching is being used to define symmetric beams and large mass block of equal thickness together in SOI device layer (up to 100 μm) in a single step to avoid alignment error in double side process. An actual accelerometer which is designed for 50 g measure range is fabricated with six lithography steps. Measurement results show 0.1166 V/g sensitivity and 0.022 % nonlinearity error in ±1 g gravity static response test. The accelerometer also provides a power spectrum less than 10.49 μVrms/Hz1/2 (89.97 μg/Hz1/2) in a non-isolated laboratory environment with a capacitive interface circuit.

A 1 × 8 MEMS (micro-electronic mechanical system(-based flexible integrated optical device with optical switch and variable optical power attenuating functions has been proposed to optimally serve optical fiber-based networking applications.The device manipulates the light with binary-slope mirrors driven by compound electrostatic actuator. The optical models for switching and attenuating are investigated, respectively. Performances of the fabricated devices assembled with ball-lensed fibers are measured and discussed. Measurements reveal that the insertion loss of the device is less than 2.8 dB and the controllable attenuation range is up to 39 dB, respectively. Moreover, polarization-dependent loss is less than 0.8 dB in the whole attenuation range. © 2012 Wiley Periodicals, Inc. Microwave Opt Technol Lett 55:231–236, 2012; View this article online at wileyonlinelibrary.com. DOI 10.1002/mop.27293

In order to reduce the computational complexity of kernel machines and improve their performance in multi-label classification, we develop a systematic two step batch approach for constructing and training a new multiclass kernel machine (MKM). The proposed paradigm prunes the kernels, and uses Newton's method to improve the kernel parameters. In each iteration, output weights are found using orthogonal least squares. Algorithm performance is compared to that of least square support vector machines and support vector machines. Simulation results on many benchmark and real life datasets show that the proposed algorithm results in smaller network size and better generalization.

This paper reports the design and theoretical calculation of a capacitive ultra-low pressure sensor, which is based on circular suspend graphene diaphragm array. The atom scale thickness and high mechanical strength of suspend graphene diaphragm both contribute to ultra-low pressure measure range. As the size of single suspend graphene diaphragm is limited by fabrication process, a parallel connected sensor array is designed to achieve detectable capacitance change and redundancy. In calculation, a sensor array with 40,000 5µm radius suspend graphene diaphragm cells in 4mm×4mm size can provide 288fF/Pa sensitivity to pressure load.

This paper reports a novel shock protector on the basis of compliant latching system for the shock resistance enhancement of microelectromechanical systems (MEMS) devices applied in high shock environment. For the first time, MEMS latches have been employed to dissipate energy during impact. Through the machine logic, the MEMS latches have realized different shock protection functions upon relative impact amplitudes to reduce the impact force applied on microstructures. In this work, the compliant latching stopper has been verified to provide multi-fold shock resistance improvement compared with some traditional shock protection structure such as hard stopper and compliant spring stopper.

In this paper, we demonstrate a testing method for characterization of microgyroscopes using a virtual rate-table, which uses a series of voltage signals to emulate the Coriolis force induced by the angular rate inputs to obtain the frequency response of the gyroscope. The proposed approach holds the following advantages: 1) it provides a convenient and efficient way to evaluate the scale factor and bandwidth of the gyroscope operating in either open-loop mode or closed-loop mode, the laborious debugging by frequently utilizing the real rate-table can be avoided; 2) it can easily identify the dynamic response to external angular rate, which is the control plant during the rebalance control design for the sense mode in a large frequency range, avoiding the performance limit of the ordinary rate-table; and 3) it can be used in a self-test for the microgyroscope system for the error calibration and malfunction checking. The method was applied to a decoupled z -axis gyroscope. The test results show that the measured scale factor and bandwidth are 30.2 mV/(deg/s) and 8.0 Hz by the virtual rate-table method, which are in close agreement with the conventional rate-table method, i.e., 31.0 mV/(deg/s) and 7.4 Hz. The static calibration with the virtual rate-table was also evaluated. The scale factors measured with the conventional rate-table method and the virtual rate-table method are 31.0 mV/(deg/s) and 30.1 mV/(deg/s) with R2 nonlinearity of 0.03% and 0.02%, respectively.

This paper presents a design method of force rebalance control for the sense mode of a micromachined vibratory gyroscope with multiobjective. The control strategy is mainly based on constraining sensitivity margin specifications via numerical optimization approach, which embeds the sense mode in a closed-loop system to possess more robust performance. This paper also provides a quantitative methodology of configuring the system parameters to realize the functional electrostatic force feedback control for the microgyroscope in the case of nonideal coupling and external angular disturbance. Theoretical results predicted by the control loop are shown to be in close agreement with the experimental results using a practical microgyroscope, which indicated a satisfactory performance of the proposed control algorithm. The maximums of the sensitivity function and complementary sensitivity function are measured to be 4.5 and 1 dB, resulting in GM ≥ 7.84 dB and PM ≥ 35°. The gyroscope achieves a scale factor of 7.1 mV/deg/s with nonlinearity of 0.2% which is improved by one order of magnitude compared with that of the open loop. The bandwidth is extended to about 98 Hz from 30 Hz in the open loop. The quadrature error and temperature stability of the scale factor are reduced from - 10.33 to -59.22 dB and from 4858 to 646 ppm/°C with the force rebalance control, respectively.

In this letter, we investigated the behaviors of surface- and buffer-induced current collapse in AlGaN/GaN high-electron mobility transistors (HEMTs) using a soft-switched pulsed \(I-V\) measurement with different quiescent bias points. It is found that the surface- and buffer-related current collapse have different relationship with the gate and drain biases ( V \(_{\mathrm { {\!GS0,}}}\) V \(_{\mathrm { {\!DS0}}}\) ) during quiescent bias stress. The surface-induced current collapse in devices without passivation monotonically increases with the negative V \(_{\mathrm { {\!GS0}}}\) , suggesting that an electron injection to the surface from gate leakage is the dominant mechanism and the Si3N4 passivation could effectively eliminate such current collapse. The buffer-induced current collapse in devices with intentionally carbon-doped buffer layer exhibits a different relationship with V \(_{\mathrm { {\!GS0}}}\) after surface passivation. The buffer-related current collapse shows a bell-shaped behavior with V \(_{\mathrm { {\!GS0}}}\) , suggesting that a hot electron trapping in the buffer is the dominant mechanism. The soft-switched pulsed \(I-V\) measurement provides an effective method to distinguish between the surface- and buffer-related current collapse in group III-nitride HEMTs.

A new simulation algorithm based on density functional theory is proposed in the application of the simulation of heterostructure HEMT device. Gauss bias is adopted and new energy functional is developed to solve the Schrodinger and Poisson equations in the HEMT device. This algorithm can avoid large data storage and reduce the simulation time of HEMT device simulation. One dimension case with nonuniform mesh shows the advantage of this new algorithm.Deducing from density functional theory, this new simulation algorithim is suitable for a full 3D simulation of semiconductor device.

Direct oxidation of composite Ni/Ti metal film structure for AlGaN/GaN MOSHEMT has been successfully demonstrated. In comparison with normal HEMT with Schottky-gate, transistors fabricated with this novel process exhibit three orders of magnitude reduction in gate leakage current, superior breakdown voltage (Vbr = 471 V vs. 88 V for normal HEMT) and electrical stability (∼0.3% electric field stress induced drain current degradation versus ∼6% for normal HEMT after 25 V drain bias). The drastic improvement in device performance stability, renders the new process highly promising for GaN based, microwave power amplifier applications in communication and radar systems.

We report the effect of the gate recess process and the surface of as-etched GaN on the gate oxide quality and first reveal the correlation between border traps and exposed surface properties in normally-off Al2O3/GaN MOSFET. The inductively coupled plasma (ICP) dry etching gate recess with large damage presents a rough and active surface that is prone to form detrimental GaxO validated by atomic force microscopy and X-ray photoelectron spectroscopy. Lower drain current noise spectral density of the 1/f form and less dispersive ac transconductance are observed in GaN MOSFETs fabricated with oxygen assisted wet etching compared with devices based on ICP dry etching. One decade lower density of border traps is extracted in devices with wet etching according to the carrier number fluctuation model, which is consistent with the result from the ac transconductance method. Both methods show that the density of border traps is skewed towards the interface, indicating that GaxO is of higher trap density than the bulk gate oxide. GaxO located close to the interface is the major location of border traps. The damage-free oxidation assisted wet etching gate recess technique presents a relatively smooth and stable surface, resulting in lower border trap density, which would lead to better MOS channel quality and improved device reliability.

A flexible integration of optical switch with optical power splitting and attenuating functions has been proposed to optimally serve optical fiber-based networking applications. In this switch, an etched binary-slope sidewall reflector is electrostatically actuated in and out of the plane to manipulate optical signals between input and output optical fibers. The fabrication process is a simple combination of a bulk-silicon micromachining process and silicon-to-glass anodic bonding where fiber alignment grooves, reflectors and actuators are fabricated on the same silicon substrate. Ball-lensed fibers are assembled with the device to achieve high coupling efficiency. Performances of the fabricated devices are measured and discussed. The switching time is less than 9 ms at 31 V. The excess loss of the device is less than 3 dB and the controllable attenuation range is up to 38 dB at 139 V, respectively. Moreover, polarization-dependent loss is less than 0.7 dB in the whole attenuation and splitting range.

This letter presents a novel MEMS-based multifunction-integrated optical device capable of performing as variable optical power splitter (VOPS), optical switch (OS), and variable optical attenuator (VOA). The device manipulates the light with a binary-slope mirror driven by compound electrostatic actuator. The optical models for splitting and attenuating are investigated, respectively. Ball-lensed fibers are assembled with the device to achieve high coupling efficiency in this work. Measurements reveal that the excess loss of the device is less than 3 dB, and the controllable attenuation range is up to 39 dB. Moreover, polarization-dependent loss is less than 0.7 dB. © 2008 Wiley Periodicals, Inc. Microwave Opt Technol Lett 50: 2185–2189, 2008; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/mop.23603

We report on a novel micro torsion mirror actuated by a compound electrostatic driving structure. The compound driving structure is made up of a planar plate drive and a vertical comb drive. The device is fabricated in SOI substrate by using a bulk-and-surface mixed silicon micromachining process. As demonstrated by experiment, the novel driving structure can actuate the mirror to achieve large-range continuous rotation as well as spontaneous 90° rotation induced by the pull-in effect in a voltage control fashion. The continuous rotating range of the micro torsion mirror is increased to about 46° at an increased threshold voltage. The measured threshold voltage ranges of the mirrors with torsion springs of 1 μm and 0.5 μm in thickness are 390–410 V and 140–160 V, respectively. Theoretical modeling is carried out to compare with the experiment. The optical insertion loss has also been measured to be −1.98 dB when the mirrors serve as digital optical switches in a 2 × 2 array form.

This paper presents a novel atmospheric electric field sensor based on high sensitivity resonant MEMS (microelectromechanical systems) electric field sense chips, which resolves the problems of motor wear, high power consumption and failure rate of the conventional electric field mill. The chip was fabricated in a commercial SOIMUMPS process. The sensor overall structure scheme was proposed and the phase-sensitive detection method was used for extract the chip output signal. A special protective structure was designed for protecting the sensor against the influences of the external environment. The minimum detectable electric field of the sensor is 10V/m with an uncertainty of 0.67% in the range of 0∼50kV/m, and its power consumption is only 0.62w. The outdoor test showed that the plotted data of the sensor agreed well with those of the conventional electric field mill on both sunny day and thunderstorm day, which indicates that the developed sensor can meet the applications of lighting hazard warning.

We report a coupling mode transition from pure plasmon coupling to plasmon-microcavity hybridization in THz using liquid metal whose shape can be tuned, not only from disk to droplet, but also from 2D to 3D.

This paper proposed a new method to classify transient power quality disturbances using S-transform and K-Nearest Neighbor classifier. First, S-transform was applied to perform time-frequency analysis on the power quality disturbance signals, and the S-transform mode time-frequency matrix was achieved from the S-transform, and we got 8 valid features extracted from the matrix, then the feature vectors who consisted of the former features were input into the Fuzzy KNN classifier to perform the disturbing signals intelligent classification, then completed the intelligent classification for six types of transient power quality disturbances signals. The result showed that the structure of the proposed method was simple, the number of the needed features was little and it was doing well in classifying these types of transient power quality disturbances.