Search

We demonstrate the transfer-printing-based integration of III-V substrate-illuminated p-i-n photodetectors onto a 4-wavelength-channel silicon photonics polarization diversity receiver circuit. $4\times 25$ Gbps operation is demonstrated for 2.5-nm wavelength channel spacing in the ${C}$ -band. Transfer printing is an enabling technology for massively parallel heterogeneous III-V-on-Si integration. [ABSTRACT FROM AUTHOR]

We demonstrate coupled bulk and surface acoustic wave (BAW/SAW) resonators with significantly improved coupling efficiency as compared to SAW only devices. Two sets of stacks are investigated: one uses a thin film piezoelectric material with a bottom electrode on GaN substrate (i.e., AlN/Mo/GaN) and another includes a piezoelectric thin film with bottom metal on a Si substrate (i.e., AlN/Mo/Si). A vibrating BAW transducer induces a SAW in the substrate and between two sets of interdigitated transducers improving the coupling efficiency by a factor of ${\geq }\, \mathsf {8}{x}$ as compared to the conventional SAW. The performance of the coupled BAW/SAW resonators using each of the Si and GaN substrates is compared using FEM analysis. By using FEM, coupling efficiency of the hybrid BAW/SAW with AlN/Mo/Si stack is 2.4%, whereas it is 3.2% for the SAW using AlN/Mo/GaN stack. [ABSTRACT FROM AUTHOR]

In this paper, we demonstrate the fabrication of contour-mode resonators (CMRs) with Al0.83Sc0.17N as a piezoelectric layer. Moreover, we assess the electromechanical coupling and the maximum achieved quality factor from 150 to 500 MHz. In comparison to pure aluminum nitride (AlN) CMRs, our results show electromechanical coupling coefficients of more than a $2\times $ factor higher at around 200 MHz. The highest quality factor is measured on a CMR operating at 388 MHz and is in excess of 1600. From the characterization of devices operating at different frequencies, material parameters of the Al0.83Sc0.17N are extracted such as the stiffness constant, the relative permittivity, and the piezoelectric constant. In particular, the reported d31 piezoelectric constant is equal to −3.9 pm/V. This represents a $2.25\times $ improvement when compared to pure AlN. Finally, we report the first temperature compensation experimental results for Al0.83Sc0.17N CMRs. Our results show that about 1.5 $\mu \text{m}$ of sputtered oxide, deposited on top of a released resonator, allows near zero temperature coefficient of frequency variation for CMRs operating up to 500 MHz. [ABSTRACT FROM AUTHOR]

Lateral Pd/n-AlN Schottky barrier diodes (SBDs) were fabricated and subjected to 3-MeV proton irradiation at various fluences. Electrical and material characterization analysis was performed before and after each radiation fluence to quantify the change in device characteristics. It was found that the SBDs performed reliably up to a proton irradiation fluence of $5 \times 10^{13}~\text{cm}^{-2}$ , with little or no change in the key device performance, such as current, turn-on voltage, ideality factor, breakdown voltage, and so on. The electrical characteristics of the SBDs were well-predicted using a standard thermionic emission theory. The performance of the SBDs shows a significant degradation after a high-fluence irritation of $5 \times 10^{15}~\text{cm}^{-2}$ , where the current of the SBDs dropped two orders in magnitude. Material and surface characterizations, including atomic force microscopy and X-ray diffraction, indicated a consistent degradation in the AlN bulk crystal quality and a drastic increase in surface roughness. These results provide valuable information on the radiation properties of AlN electronics and can serve as important references for the future development of high-performance AlN devices for extreme environment applications. [ABSTRACT FROM AUTHOR]

The pulsed-laser single-event transient response of InGaAs FinFETs on bulk silicon substrates is investigated. Charge collection due to a source–drain shunt effect and drain-to-substrate junction charge collection contribute to the observed transients. The transient response of these silicon substrate devices is compared to that of InGaAs FinFETs on semi-insulating substrates. Faster transients with reduced peak currents and peak widths are observed on the silicon substrate devices. Simulations show hole collection by the silicon substrate. This reduces the amount of source-barrier lowering and bipolar-amplification relative to other III–V devices. Moreover, the reduced hole lifetime in the GaAs buffer layer also contributes to the relative reduction of the bipolar amplification in these devices. [ABSTRACT FROM AUTHOR]

We report on an on–off keying intensity-modulation and direct-detection C-band optical transceiver capable of addressing all datacenter interconnect environments at well beyond 100 Gbaud. For this, the transmitter makes the use of two key InP technologies: a 2:1 double heterojunction bipolar transistor selector multiplexer and a monolithically integrated distributed-feedback laser traveling-wave electro-absorption modulator, both exceeding 100-GHz of 3-dB analog bandwidth. A preamplified 110-GHz PIN photodiode prior to a 100-GHz analog-to-digital converter complete the ultrahigh bandwidth transceiver module; the device under study. In the experimental work, which discriminates between intra- and inter-data center scenarios (dispersion unmanaged 120, 560, and 960 m; and dispersion-managed 10 and 80 km of standard single-mode fiber), we evaluate the bit-error rate evolution against the received optical power at 140, 180, and 204 Gbaud on–off keying for different equalization configurations (adaptive linear filter with and without the help of short-memory sequence estimation) and forward error correction schemes (hard-decision codes with 7% and 20% overhead); drawing conclusions from the observed system-level limitations of the respective environments at this ultrahigh baudrate, as well as from the operation margins and sensitivity metrics. From the demonstration, we highlight three results: successful operation with >6-dB sensitivity margin below the 7% error-correction at 140 Gbaud over the entire 100 m–80 km range with only linear feed-forward equalization. Then, the transmission of a 180-Gbaud on–off keying carrier over 80 km considering 20% error-correction overhead. Finally, a 10-km communication at 204 Gbaud on–off keying with up to 6 dB sensitivity margin, and regular 7% overhead error-correction. [ABSTRACT FROM AUTHOR]

We report on the first realization of InAs n-channel gate-all-around nanowire MOSFETs on 300 mm Si substrates using a fully very large-scale integration (VLSI)-compatible flow. Scaling of the equivalent oxide thickness EOT in conjunction with high-κ dielectric engineering improves the device performance; with an optimized gate stack having an EOT of 1.0 nm, the sub-threshold swing S is 76.8 mV/dec., and the peak transconductance gm is 1.65 mS/μm, at Vds of 0.5 V, for a gate-all-around nanowire MOSFET having a gate length Lg of 90 nm, a nanowire height HNW of 25 nm, and a nanowire width WNW of 20 nm, resulting in Q ≡ gm/S = 21.5, a record for InAs on silicon. Furthermore, we report a source/drain resistance Rsd of 160-200 Ω·μm, amongst the lowest values reported for III-V MOSFETs. Our VLSI-compatible process provides high device yield, which enables statistically reliable extraction of electron transport parameters, such as unidirectional thermal velocity vtx of 3-4×107 cm/s and back-scattering coefficient rc as a function of gate length.

A working small-footprint asymmetrically and vertically coupled hybrid Si/GaN microring resonator (HMR) was demonstrated. The HMR of a minimum radius of 20 μm was fabricated in a high-yield (~90%) hybrid nanophotonic platform that allowed interactions between the GaN microring and the underlying Si waveguide circuit. The HMR's spectral response across the telecommunication C- and L-bands was measured. The high extinction ratio of up to 17 dB, the resonance linewidth as narrow as 40 pm, and the intrinsic quality factor of up to 70 000, which was the highest value for GaN microring reported so far, were achieved. The explicit analytical model for the high-index-contrast HMR was developed. Our basic study opened new possibilities for exploring GaN-Si nonlinear phenomena and for developing complex on-chip optical interconnects.

Effects of source/drain (S/D) doping density (NSD) on the ballistic performance of III-V nanowire (NW) n-channel metal-oxide-semiconductor field-effect transistors (n-MOSFETs) are explored through atomistic quantum transport simulation. Different III-V materials (InAs, GaAs) and transport directions (, ) are considered with Si included for benchmarking for a gate length of 13 nm. For III-V's, depending on the operating condition (OFF-current target for a given supply voltage), there exists an optimum NSD that maximizes ON-current (ION) by balancing source exhaustion versus tunneling leakage. For InAs, sub-threshold swing degrades significantly with increasing NSD due to the light effective mass (m*) and source-drain tunneling, so the optimum NSD is low. For GaAs, such dependence is much weaker due to the larger m*, and the optimum NSD is higher. With optimized NSD's, InAs shows low ballistic ION due to the low density-of-state (DOS) whereas GaAs NW with transport direction shows good ballistic ION due to the improved DOS with still high injection velocity, making it a better candidate for high performance device.

In this paper, we introduce p-channel InAs-Si tunnel field-effect transistors (TFETs) fabricated using selective epitaxy in nanotube templates. We demonstrate the versatility of this approach, which enables III-V nanowire integration on Si substrates of any crystalline orientation. Electrical characterization of diodes and of TFETs fabricated using this method is presented; the TFETs exhibit a good overall performance with on-currents, Ion of 6 μA/μm (|VGS| = |VDS| = 1 V) and a room-temperature subthreshold swing (SS) of ~160 mV/dec over at least three orders of magnitude in current. Temperature-dependent measurements indicate that SS is limited by traps. We demonstrate improved TFET Ion performance by 1-2 orders of magnitude by scaling the equivalent oxide thickness from 2.7 to 1.5 nm. Furthermore, a novel benchmarking scheme is proposed to allow the comparison of different TFET data found in literature despite the different measurement conditions used.

The thermal performance of Fabry–Perot InP lasers integrated onto different silicon photonics substrates by microtransfer printing is assessed. 500-μm-long ridge waveguide lasers on the original 350-μm-thick InP have an experimental thermal impedance, ZEXP, of 57 K/W that is reduced to 38 K/W after printing to a 500-μm-thick Si substrate. ZEXP for lasers printed on silicon-on-insulator wafers is ∼94 K/W, which is more than two times higher than that of the laser printed on the Si substrate. ZEXP of lasers printed on thermally insulating layers like benzocyclobutene (BCB) or SiO2 increases with the thickness of the layer. BCB adhesive layers as thin as 50 nm limit ZEXP to be greater than 55 K/W. The thermal properties for the different situations were modeled using finite-element simulations which confirmed the experimental results within 10% accuracy. The simulations show how changes in the geometry and the materials of the integration platform can influence the resulting thermal impedance. [ABSTRACT FROM AUTHOR]

We describe an aluminum nitride (AlN)-based resonant switch for use in a near zero power radio frequency (RF) wake-up receiver. A folded beam structure with a slot compensates for the curvature caused by the stress gradient in the sputtered AlN film and ensures that the released contact gap is approximately equal to the designed contact gap. A 80 kHz resoswitch with a Q of 8600 and an actuation gap of approximately 600 nm turns on when a −4 dBm, 800 MHz signal, square wave modulated at 80 kHz, is applied to the actuator. This AlN electrostatic resonant switch is designed to enable integration with a high gain AlN RF piezoelectric transformer to form a complete ultra-low power RF receiver. [ABSTRACT FROM AUTHOR]

The characteristics of the multi-mode resonance behavior of AlN Lamb wave resonators (LWRs) are theoretically and experimentally investigated for the first time in this paper. Adler’s approach and finite element method (FEM) are used to calculate the dispersive characteristics of the phase velocity ($v_{p}$), group velocity ($v_{g}$), effective coupling coefficient ($k^{2}_{\text {eff}}$), and temperature coefficient of frequency for the first eight Lamb wave modes with different transducer configurations. The FEM is performed to take an insight into the mode shapes of the S0mode and S1mode specifically: the S0mode is more contour-like and exhibits the largest $k^{2}_{\text {eff}}$ when $h_{\text {AlN}}/\lambda $ is close to 0.5; the S1mode is strong in vertical direction and can enable high resonance frequency ($f_{s}$) and large $k^{2}$ simultaneously when AlN thickness is very thin. Experimentally, AlN LWRs with different AlN thicknesses are designed and fabricated. The measured results are fitted into the multi-resonance BVD model so that the device performance parameters, as well as the equivalent, lumped element values are extracted and compared. By choosing different normalized AlN thicknesses, the performance of different Lamb wave modes varies largely due to the dispersive characteristics and agrees well with theoretically predicted acoustic characteristics. This paper lays the foundation for characterizing the multi-resonance behaviors of AlN LWRs and gives guidance on choosing the optimal design parameters and Lamb wave modes for different applications. [2018-0040] [ABSTRACT FROM AUTHOR]

Current transient spectroscopy was used to measure the impact of neutron irradiation on output current-limiting charge traps in AlGaN/GaN high electron mobility transistors with time constants from 10 ms to 1800 s. We find that coupling between discrete traps was apparent, in contrast to the commonly employed assumption of independent trap (dis)charging, and increased after 14-MeV neutron irradiation of $2 \times 10^{13}\,\,\text {n/cm}^{2}$ and above. Irradiation to a high dose of as much as $7.8 \times 10^{14}\,\,\text {n/cm}^{2}$ , which is comparable to eight years exposure to a harsh radiation environment, such as the International Thermonuclear Experimental Reactor neutral-beam injector prototype, increased trapped charge density and reduced transient drain current to as little as 75% of its equilibrium value. These changes are consistent with displacement damage estimates based on the radiation transport calculations. [ABSTRACT FROM AUTHOR]

We present precision scattering-parameter measurements of chip-to-chip connections in heterogeneous integrated circuits: indium phosphide or gallium nitride “chiplets” mounted on Silicon Complementary Metal-Oxide-Semiconductor carrier chips. We demonstrate methodology, experimental results, and modeling results of these chip-scale interconnections from dc to 110 GHz. We used thru-reflect-line on-wafer calibration to establish reference planes inside heterogeneous integrated circuits, and then, we translated those reference planes to the proximity of the chip-to-chip transitions to isolate their contribution to the scattering parameters. [ABSTRACT FROM AUTHOR]

In this paper, we address the sharing incentive issue in multi-tenant passive optical networks (PONs). We propose an economic-robust and efficient sharing platform for new emerging multi-tenant PON networks. This platform is capable of accommodating a diverse range of service providers and enhancing the network utilization. We propose a sharing platform that provides sharing incentives for the incumbent network operators through monetization of inter-operator network sharing. Meanwhile, the platform allows the incumbent operators to operate a virtual instance of the bandwidth scheduling algorithm that enables them to meet their quality of service and latency requirements. Therefore, the proposed sharing platform grants a high degree of control to the operators co-operating the same network while, thanks to the higher resource efficiency, reduces the initial investment. We first model the multi-tenant PON as a market and define the roles of the virtual network operators and the infrastructure provider along with their utility functions. We propose a double auction mechanism to facilitate the trading of excess resources. The proposed double auction satisfies the crucial economic properties of a market while it achieves more efficient resource allocation among the market players. We have theoretically proven the economic robustness of the mechanism including incentive compatibility, individual rationality, and weak budget balance. Through extensive market simulations, we confirmed that the proposed mechanism achieves superior allocative efficiency compared to a reference baseline mechanism. [ABSTRACT FROM AUTHOR]

This letter reports the first results on a new generation of anti-symmetric lamb-wave devices using a simple solidly mounted platform of AlN/Mo/Si and operating at very-high-to-super-high frequency range. AlN/Mo/Si-based solidly mounted resonators (SMRs) with sub-micrometer interdigitated transducers were fabricated using the electron beam lithography and liftoff process. Plasma etching was used to etch vias in ground pads to get access to the bottom electrode for grounding. By comparing the floating, grounded, and no bottom electrode device performances, it has been observed that the presented SMR design with grounded bottom electrode results in enhancement of the coupling efficiency by 500% compared with conventional SAW devices. In addition, these devices are more mechanically robust and can be monolithically integrated with a CMOS platform without requiring complex packaging processes compared with conventional released lamb-wave resonators. [ABSTRACT FROM AUTHOR]

In this letter, we investigated the impact of the source-to-drain spacing (${L}_{\textsf {SD}}$) on the dc and high-frequency characteristics of indium-rich In0.8Ga0.2As/In0.52Al0.48As high-electron mobility transistors (HEMTs) on a 3-in InP substrate. ${L}_{g} = \textsf {87}$ nm HEMTs with different values of ${L}_{\textsf {SD}}$ were fabricated ranging from 1.55 to $0.8~\mu \text{m}$ , and their dc and RF responses were measured and analyzed in detail. In order to suppress the increase of the parasitic gate capacitance, we maintained the gate stem height as high as 200 nm in our device fabrication. Both the maximum transconductance (gm_max) and on-resistance (${R}_{\textsf {ON}}$) improved as ${L}_{\textsf {SD}}$ scaled down to $0.8~\mu \text{m}$. At the same time, the high-frequency figures of the merit, such as current-gain cutoff frequency (${f}_{T}$) and maximum oscillation frequency (${f}_{\textsf {max}}$), increased with the reduction of ${L}_{\textsf {SD}}$. These improvements are attributed to the reduction of series resistances. In particular, the ${L}_{g} =87$ nm device with an ${L}_{\textsf {SD}} = \textsf {0.8} \mu \text{m}$ exhibited an excellent combination of dc and RF characteristics, such as gm_max= 2.7 mS/ $\mu \text{m}$ , ${R}_{\textsf {ON}} = \textsf {318}\,\,\Omega \cdot \mu \text{m}$ , ${f}_{T} = \textsf {519}$ GHz, and ${f}_{\textsf {max}} = \textsf {645}$ GHz, respectively. The results obtained in this letter indicate that the reduction of ${L}_{\textsf {SD}}$ down to $0.8~\mu \text{m}$ continues to improve both the dc and RF characteristics of the InGaAs/InAlAs HEMTs, and a possible increase in parasitic capacitance components, associated with a T-shaped, is still negligible in our device architecture. [ABSTRACT FROM AUTHOR]

In this brief, an AlGaN/GaN metal–insulator–semiconductor high-electron-mobility transistor (MIS-HEMT) with pulsed laser deposited AlN interface protection layer (IPL) and trench termination (${T}^{\textsf {2}}$) structure is experimentally and theoretically investigated. The AlN IPL could effectively improve the interface quality and reduce the interface trap density, which is verified by frequency-dependent ${C}$ – ${V}$ measurement and conductance method. The ${T}^{\textsf {2}}$ structure extends the depletion region and increases the average electric field (E-field) strength between the gate and drain, achieving an enhanced breakdown voltage (BV). The measured BV and saturated output current density are 412 V and 505 mA/mm for the device at ${L}_{\textsf {G}}/{L}_{\textsf {GS}}/{L}_{\textsf {GD}}/{W}_{\textsf {G}}= \textsf {1.5}$ /1.5/5/ $10~\mu \text{m}$ , respectively. Compared with the MIS-HEMT without ${T}^{\textsf {2}}$ structure, the proposed device increases the BV by 63%. [ABSTRACT FROM AUTHOR]

We present two monolithic microwave integrated circuit (MMIC) cryogenic broadband low-noise amplifiers (LNAs) based on the 100 nm gate length InP high-electron mobility transistor technology for the frequency range of 0.3–14 and 16–28 GHz. The 0.3–14 GHz three-stage LNA exhibited a gain of 41.6 ± 1.4 dB and an average noise temperature of 3.5 K with a minimum noise temperature of 2.2 K at 6 GHz when cooled down to 4 K. The 16–28 GHz three-stage LNA showed a gain of 32.3 ± 1.8 dB and an average noise temperature of 6.3 K with a minimum noise temperature of 4.8 K at 20.8 GHz at the ambient temperature of 4 K. This is the first demonstration of cryogenic MMIC LNA covering the whole K-band. To the best of the authors’ knowledge, the cryogenic MMIC LNAs demonstrated the state-of-the-art noise performance in the 0.3–14 and 16–28 GHz frequency range. [ABSTRACT FROM AUTHOR]

The massive growth of data traffic generated by human and machine activity has significantly increased the demand for photonics products for optical communications. In order to meet the current market demand at lower price points, it is necessary to increase the production volume and the wafer size for printing more devices. For this reason, high-volume manufacturing processes for lasers and photodiodes were developed using the existing equipment in a GaAs IC wafer fabrication facility, without a major capital investment. A variety of highly reliable, high-performance directly modulated edge emitting lasers and front illuminated photodiodes were manufactured on 4 inch InP substrates. [ABSTRACT FROM AUTHOR]

Monolithic and hybrid integration platforms are both important for photonic integrated circuits. It is desirable for the laser sources integrated in both platforms to provide high power and single frequency operation for many emerging applications. However, it is difficult to realize simultaneous high power and single frequency operation by using conventional semiconductor distributed feedback lasers. Integrated coherent beam combining (CBC) is a promising solution to overcome this challenge. In this paper, coherently combined, integrated diode laser systems are experimentally demonstrated through both monolithic and hybrid CBC. For monolithic CBC of broad-area lasers in the InP platform, two single triangular-lattice photonic crystal (PC) Bragg lasers are combined. The measurement results show that the output power of the combined PC Bragg lasers is increased and the single frequency operation is maintained. As for the hybrid CBC, a chipscale coherently combined laser system is experimentally demonstrated in the InP-Si3 N4 platform through the manipulation of optical feedback at different output ports of the coupled laser cavities. Coherent combining of two InP-based reflective semiconductor amplifiers are obtained by use of the cross-coupling provided by an adiabatic 3-dB coupler in silicon nitride, with a combining efficiency of ∼92%. [ABSTRACT FROM AUTHOR]

To realize cost-effective and adaptive network control and management (NC&M) on inter-datacenter optical networks (IDCONs), people have considered network virtualization to let the operator of an IDCON work as an infrastructure provider (InP), which can create virtual optical networks (VONs) over the IDCON for tenants. In this paper, we use this network scenario as the background, and try to integrate deep learning (DL) based traffic prediction in the NC&M of the IDCON and the VONs created over it. We first design the service provisioning framework in which each tenant uses a DL module to predict the traffic in its VON and will submit a VON reconfiguration request to the InP, when it sees a significant mismatch between future traffic and the allocated resources in its VON. Then, the InP will invoke the VON reconfiguration to make the VON be better prepared for future traffic. An adaptive and scalable DL-based traffic predictor is proposed together with a cognitive service provisioning algorithm to exploit the temporal and spatial characteristics of interDC traffic and achieve effective service provisioning based on precise and timely traffic prediction. Next, we consider the situation where a tenant leverages “machine-learning-as-a-service” and outsources the training of its DL module to a third-party entity for overcoming its resource limitations, and analyze the induced vulnerabilities due to data poisoning. Our simulation results indicate that with our proposal, the InP can invoke VON reconfigurations timely and improve the service provisioning performance of each VON significantly. Meanwhile, the results also demonstrate that our data poisoning scheme can easily bypass the normal validation of the DL module and generate significant adversarial effects. [ABSTRACT FROM AUTHOR]

This paper presents a novel method to change the dielectric constant and enhance the average power-handling capability (APHC) to a thin-film coplanar waveguide (CPW) line through dc reactive magnetron sputtering. This paper discusses the effects of material properties, microwave characteristics, and APHC on the thin-film CPW line. An aluminum nitride (AlN) thin film can be deposited on the surface of a ceramic substrate to provide a substrate with a smooth surface. Using the proposed thin-film passivation on the ceramic substrate can effectively improve the thermal conductivity and enhance the APHC of a thin-film CPW line. In a physical experiment, a thin-film CPW line with a 3- $\mu \text{m}$ -thick AlN thin film was deposited on a 1000- $\mu \text{m}$ -thick ceramic substrate. Microwave losses (including conductor loss $\alpha _{c}(f)$ and dielectric loss $\alpha _{d}(f)$) dielectric constant and APHC were extracted from S-parameters that were measured up to 10 GHz. The method can be applied in ceramic fabrication processes, particularly in low-temperature cofired ceramic techniques. [ABSTRACT FROM AUTHOR]

In this paper, the phase noise of aluminum nitride (AlN) contour-mode resonators is investigated using a passive measurement system with carrier suppression. The purpose is to make careful measurements of the performance of AlN resonators in order to better understand and clarify previously reported frequency instability in these devices. The resonant frequencies of the resonators are around 220 MHz. The motional parameters, the thermal behavior, and the nonlinear power effect of these resonators have been evaluated. Then, the principle of the noise measurement system is reviewed, and the resonator conditioning is shown. Finally, the noise measurements of the resonators are presented and discussed. [ABSTRACT FROM AUTHOR]

Surface acoustic wave sensors find their application in a growing number of fields. This interest stems in particular from their passive nature and the possibility of remote interrogation. Still, the sensor package, due to its size, remains an obstacle for some applications. In this regard, packageless solutions are very promising. This paper describes the potential of the AlN/ZnO/LiNbO3 structure for packageless acoustic wave sensors. This structure, based on the waveguided acoustic wave principle, is studied numerically and experimentally. According to the COMSOL simulations, a wave, whose particle displacement is similar to a Rayleigh wave, is confined within the structure when the AlN film is thick enough. This result is confirmed by comprehensive experimental tests, thus proving the potential of this structure for packageless applications, notably temperature sensing. [ABSTRACT FROM AUTHOR]

Network virtualization (NV) allows multiple heterogeneous virtual networks (VNs) to coexist and operate over the same physical network (PN) infrastructures. Some of the benefits of this advancement include flexibility in VN topologies, heterogeneity in VN technologies, and modularity of network operations. However, there are a few areas, such as resource allocation and economics, which challenge the implementation of NV. In this paper, we first introduce some NV parameters that influence the resource allocation and economics of an NV system. Next, we formulate an economic model for NV using the prey-predator food chain model. This model takes into account the dynamics in an NV system, such as the service, payoff, failure, and competition rates within each VN and PN. The solution point to this model represents the resource strategy of the service provider (SP) given the number of users trying to use its VN, as well as the resource strategy of the infrastructure provider (InP) given the strategy of the VN leasing its PN. In addition, we establish economic models that relate the capacities of the end users, the SP, and the InP. Finally, we provided simulations that show how the prey-predator food chain model fits well on an NV system. [ABSTRACT FROM AUTHOR]

We propose an algorithm for the cloud and bandwidth resource allocation in multi-provider NFV environments. The resources are allocated so as to take into account the different costs charged by the cloud infrastructure providers (InP). The effectiveness of the proposed algorithm is confirmed from the comparison with the results of the optimal problem. Its application in medium and large networks has shown that it can lead to cost saving as high as 65% with respect to algorithms that allocate resources without taking into account the cost differences charged by the InPs. [ABSTRACT FROM AUTHOR]

In this paper, we investigate the temperature cross-sensitivity of aluminum nitride (AlN)-based flexural plate wave devices for sensing applications in contact with liquids. In our improved device topology, the interdigital transducers are designed as a buried electrode, and thus, are electrically shielded, enabling full immersion of the sensor into the liquid. The fabricated devices showed a mass sensitivity of 240 cm2/g for loading with deionized water which is in good agreement with the theoretical predictions. The temperature coefficient of frequency (TCF) was evaluated for devices with initially compressive and tensile stressed AlN layers. Devices with compressive film stress exhibited a TCF of −62 ppm/K to −28 ppm/K, while the devices with tensile film stress showed an increased TCF of −391 ppm/K to −72 ppm/K. It is shown that variations in in-plane tension are mainly accountable for the increased TCF, while the lower TCF is primarily caused by material softening. [ABSTRACT FROM AUTHOR]

In this brief, an innovative terahertz (THz) ballistic deflection transistor (BDT) travelling wave amplifier (TWA) system is proposed. The BDT is an innovated technology device based on InGaAs/InAlAs/InP, able to operate at THz frequencies. The whole BDT TWA system design is based on a lossless THz transmission line called parallel plate dielectric waveguide with signal line (PPDWS). Parasitics for a single stage BDT amplifier and multiple-stage amplifier are also studied. In this brief a new thinner lossless PPDWS (transmission loss = 1.1 dB/mm) is developed. A new THz coupler applying the new PPDWS topology is also proposed and studied. It is shown that a 48-stage TWA system is able to provide a 15-dB gain at more than 1-THz frequency. Results compensate the low transconductance of the BDT and make the TWA a good choice for future THz amplifiers. [ABSTRACT FROM AUTHOR]

The time-dependent photocurrent response in semi-insulating GaAs and InP was studied based on 1-D, time-dependent simulations with a focus on the Lock-On phenomenon. The results underscore the role of trap-to-band impact ionization from deep traps in rapid charge creation and its subsequent propagation much like a streamer. The numerical results compare well with the actual data. The main findings are that deeper traps nearer the valence band at higher densities, materials with larger high-field drift velocity, and cathode-side illumination would all aid in attaining Lock-On. These could be useful guidelines for producing Lock-On in new materials such as GaN for high-power applications. [ABSTRACT FROM AUTHOR]

We report on the realization mymargin of transferred-substrate InP/GaAsSb double heterostructure bipolar transistors in a terahertz monolithic integrated circuit process. Transistors with 0.4- $\mu \text{m}$ -wide single emitters reached unilateral gain cutoff frequencies of around 530 GHz with simultaneous current gain cutoff frequencies above 350 GHz. Extrinsic collector capacitance is effectively reduced in the transfer-substrate process. In combination with the high collector breakdown voltage in the InP/GaAsSb heterobipolar transistor structure of 5 V, this process is amenable to analog power applications at millimeter (mm-wave) and sub-mm-wave frequencies. We demonstrate reliable extraction procedures for unilateral gain and current gain cutoff frequencies. [ABSTRACT FROM AUTHOR]

InAlAs digital alloy avalanche photodiodes exhibit lower excess noise than those fabricated from conventional random alloy material. Experiment and Monte Carlo simulation both show that relative to the random alloy the ionization probability for electrons is slightly lower while that of holes is greatly suppressed. We propose that the suppression of carrier ionization probability in digital alloys happens because of the creation of minibands that localize carriers. The difference of suppression between conduction bands and valence bands comes from the difference of scattering path. [ABSTRACT FROM AUTHOR]

In this paper, we present an ultrasound nondestructive testing (NDT) method for high–temperature (HT) applications using laser-generated Lamb waves and aluminum nitride (AlN) sensors. Lamb waves were introduced to a stainless steel plate by the Nd:YAG pulsed laser at one point and detected by an AlN receiver at a distant position. The fundamental symmetric ( $S_{0}$ ) and antisymmetric ( $A_{0}$ ) Lamb waves generated by the pulsed laser were successfully detected by the AlN sensor. This detection was done on a stainless steel plate at temperatures ranging from room temperature to about 800 °C, with a signal-to-noise ratio of higher than 20 dB. Based on the time-of-flights analysis, the ability of this NDT method to localize the defect at an HT (~800 °C) has been demonstrated. [ABSTRACT FROM AUTHOR]

In this paper, a structure of super-lattice structure last barrier (SLSLB) is proposed, which can be applied into the AlGaN-based ultraviolet light-emitting diodes (LEDs) for improving the injection of both electrons and holes. Several other SLSLBs are also designed and compared in this work, and according to the simulated results, we find that the LED with SLSLB of diminishing thickness (SLSLB-D) possesses the highest internal quantum efficiency and the smallest efficiency droop. The SLSLB-D can effectively reduce the electron concentration at the interface between the last barrier (LB) and electron block layer (EBL), which relieves the leakage of electrons. Moreover, the best hole injection capability for the LED with SLSLB-D also contributes to the improved optical performance. [ABSTRACT FROM AUTHOR]

We report a “Type-II” metal–organic chemical vapor deposition (MOCVD)-grown compositionally graded quaternary base InP/GaInAsSb/InP double heterojunction bipolar transistor (DHBT) with simultaneous cutoff frequencies ${f}_{\text {T}}/{f} _{\text {MAX}}$ = 547/784 GHz. Compositional grading of the As/In mole fractions enables stronger quasi-electric fields in the base and favors better RF performance. Consideration of the higher bandstructure reveals new insights on the performance limits of “Type-II” DHBTs: Indium incorporation in a GaAsSb base appears to increase the efficiency of electron injection from the base into the InP collector. The present results correspond to the fastest MOCVD-grown quaternary base DHBTs reported to date for a device with ${f} _{\text {MAX}}>{f}_{\text {T}}$. [ABSTRACT FROM AUTHOR]

A systematic analytical investigation of the charge distribution and gate control of the normally-off GaN double-channel MOS-HEMT (DC-MOS-HEMT) is presented in this paper. Compared to conventional GaN MOS-HEMT, the DC-MOS-HEMT features a thin AlN insertion layer (AlN-ISL) below the original two dimensional electron gas (2DEG) channel, thus forming a second channel at the interface between AlN-ISL and the underlying GaN. This paper reveals the impact of the AlN-ISL on the 2DEG distribution and the gate control of the channels. The sensitivity of ${V}_{\text {th}}$ against the recess depth is also analytically studied and is found to be nearly independent of the recess depth as long as the recess is terminated in the upper channel layer. The analytical results are well supported by numerical device simulations, and the physical mechanisms behind these findings are explained along with the analytical investigations. [ABSTRACT FROM AUTHOR]

Low-excess-loss arrayed waveguide gratings are enabled by unique application of deep UV lithography in InP integrated photonics through reduced feature sizes and, more specifically, well-resolved inter-waveguide gap dimensions. Submicrometer wafer-flatness is shown to be required to achieve the critical dimension uniformity better than 10 nm on 3-in substrates. Arrayed waveguide grating devices were fabricated and the effect of inter-waveguide gap scaling on the excess losses was measured and compared to simulations. Excess losses down to 0.15 dB were demonstrated to be lower than predicted with the 2-D simulations. The tapering of the etch depth inside the gaps due to the lag effect of the etch process may explain the improvements. [ABSTRACT FROM AUTHOR]

InP/InGaAsSb double heterojunction bipolar transistors (DHBTs) with an Au subcollector were fabricated on a SiC substrate by using a wafer bonding technique. The thermal resistance ( R\textsf {th} ) of a fabricated DHBT with a \textsf 0.35 \times \textsf 8.0\,\,\mu \textm emitter is 62% lower than that of a DHBT with the same epitaxial layer structure on an InP substrate. Thanks to the lower R\textsf {th} , the DHBT operates at a collector current density ( JC ) of over 25 mA/ \mu \textm^\textsf 2 without any lowering of current gain. The DHBT exhibits a record ft of 695 GHz at JC of 21 mA/ \mu \textm^\textsf 2 owing to the reduced emitter charging time by suppressing the Kirk effect. [ABSTRACT FROM PUBLISHER]

To get a better insight into the vertical leakage mechanism of GaN-on-Si, the carrier transport from n+, n, p+, and p-Si(111) substrates through the AlN nucleation layer was investigated. A plateau in the current-voltage curve was found only for the AlN/p-Si heterojunction due to depletion of the p-Si substrate. Detailed study illustrated that it was the leaky AlN that cannot effectively block the increasing amount of electrons in the inversion layer at the interface and triggered the depletion. Temperature-dependent characterization suggested that the forward vertical leakage mechanism of AlN/Si could be explained sequentially by Ohm’s law, space-charge-limited conduction, variable-range hopping, and trap-assisted tunneling. A model involving shallow donor traps, interface traps, and deep level traps was proposed to explain the leakage characteristics. This paper shows that the carrier concentration of the Si substrates strongly impacts the vertical leakage characteristics, and also that the carrier transport from the Si substrate through the AlN nucleation layer is heavily influenced by traps. [ABSTRACT FROM AUTHOR]

In this paper, the effects of aluminum nitride and alumina ceramics on the discharge mode of He/O2 mixed gas are studied and compared. It is found that when the O2 volume fraction is in the range of 0.2%–1%, dielectric barrier discharge (DBD) with Al2O3 is always in the unstable filamentary mode, and homogeneous and filamentary modes are observed in DBD with AlN. Under the homogeneous discharge mode, there is only one current pulse each half-voltage cycle, and the intensified charge-coupled device discharge image shows that it is glow discharge. The relative permittivity and surface charge decay characteristics of AlN and Al2O3 are measured. The experimental results show that the relative permittivity of AlN is lower than that of Al2O3, but its surface charge decay is obviously slower than that of Al2O3. The derived electron trap density from the surface charge decay of AlN ceramic is higher than that of Al2O3, which shows AlN has a strong charge capacity and can emit more electrons in the second half cycle of the discharge. The higher secondary electron emission coefficient can make the discharge ignition voltage lower and the formation of stable homogeneous discharge easier. This paper provides a new control method for expanding homogeneous discharge in He/O2 mixed gas. [ABSTRACT FROM AUTHOR]

A high quality factor ($Q$) and low-impedance aluminum nitride (AIN) on silicon piezoelectric resonator is reported in this paper. The piezoelectric resonator is fabricated based on the AIN-on-cavity-Silicon-on-Insulator (SOI) platform and vacuum encapsulated using the wafer-level chip-scale packaging by aluminum–germanium eutectic bonding approach. The fabricated resonator achieves a loaded Q$ of 9520 and a motional impedance of 32.64 \Omega$ at its series resonant frequency of 27.19 MHz. Thermal cycles (TC) and highly-accelerated stress test (HAST) are performed to evaluated the packaging reliability of the fabricated devices. No obvious impedance change caused by the TC and HAST is observed. The TC results indicate that the packaged devices have an equivalent life of more than 11.65 years based on the Coffin–Manson model. The piezoelectric resonator-based programmable oscillators demonstrate an overall frequency drift of \pm$ 3 ppm over the temperature range of -C to 70  ^{\circ }C thanks to the passive and active temperature compensations. With this stable frequency reference, the programmable oscillator produces a 75.01-MHz frequency output. The measured integrated phase jitter is 2.02 ps (rms) from 12 to 10 MHz. [ABSTRACT FROM AUTHOR]

Inline phase-change switch (IPCS) devices were fabricated on top of an AlN heat spreading layer deposited on a thick SiO2 layer. The thick SiO2 emulates the inter-layer dielectric thermal properties that prevent direct integration of IPCS devices at the end of a back-end-of-line CMOS process. The AlN acts as a heat spreading layer, enabling the proper thermal quench rate despite the low thermal conductivity of the oxide underneath. Multiple AlN thicknesses were examined and shown to properly quench the device. Use of this heat spreading layer experimentally demonstrates a substrate agnostic monolithic integration scheme, where IPCS devices can be integrated at the end of any microfabrication process—regardless of the substrate or materials underneath. This is attractive for many CMOS applications, where wafer real-estate is expensive, or in III–V applications, where the wafer thermal properties are poor and vary among different technologies. [ABSTRACT FROM AUTHOR]

We report Lg= 30 nm InAs-channel MOSFETs exhibiting 420 GHz f\textit {max} , record for a III-V MOSFET, and 357 GHz ft . The device incorporates a 5-nm strained InAs channel grown on an InP substrate. To reduce the parasitic gate-source and gate-drain capacitances, regrown lateral access regions increase the separations between the gate and the N+ source and drain; modulation doping within these access regions provides a low associated series resistance, enabling high gm . The 30 nm Lg device shows an 1.5 mS/ \mu \textm DC peak extrinsic gm at V\textit {DS} = 0.5 V and V\textit {GS} = 0.3 V, 91% of the value (1.65 mS/ \mu \textm ) extracted from 10 MHz RF measurements, indicating a low DC-RF dispersion. [ABSTRACT FROM AUTHOR]