Photoconductivity measurements in optical fibre preforms

The AC photoconductivity and the photodielectric shift in UV-illuminated germano-silicate preforms and optical fibres has been measured using a noncontact microwave resonator technique at 2.9 GHz. The highest photoconductivity measured was 1.5*10/sup -6/ ( Omega cm)/sup -1/ and the largest measured photodielectric shift corresponded to an increase of the refractive index of 1.6*10/sup -2/. A permanently induced increase of the refractive index, 6.1*10/sup -3/, was also observed. The dark DC conductivity was measured to be 10/sup -15/ ( Omega cm)/sup -1/.<>     

p-Channel MODFET's using GaAlAs/GaAs two-dimensional hole gas

p-channel MODFET's were fabricated in the GaAlAs/GaAs system and the properties of the hole gas system tested to ascertain its suitability for complementary logic. The Hall mobilities on a Ga.5Al.5As/GaAs modulation-doped hole gas structure were measured to be 3650 cm²V^-1s^-1and 54000 cm²V^-1s^-1with sheet carrier concentration of 1 × 10¹²cm^-2and 7.76 × 10¹¹cm^-2at 77 and 4.2 K, respectively. The measured transconductances of 1.5-µm gate-length MODFET's on this structure were measured to lie in the range of 28-35 mS.mm^-1at 77 K. The field mobility measured on long gate-length MODFET's was approximately 3200 cm²V^-1s^-1at 77 K. Using test structures for measuring current voltage characteristic in the hole-gas system, low field drift mobility was measured to be 3000 cm²V^-1s^-1and velocities of 3 5 × 10⁶cm.s^-1were measured at electric fields of 3-4 kV.cm^-1at 77 K. The Schottky barriers showed low leakage and a barrier height of 0.7 to 0.8 eV. Calculations indicate that transconductances of approximately 100 mS.mm^-1should be achievable in this system for similar gate lengths.     

A novel, bias-dependent, small-signal model of the dual-gate MESFET

A dual-gate MESFET from NEC (NE25000) has been measured and modeled. S-parameters and drain-to-source currents calculated from the model are in good agreement with measured data. The model consists of a cascode of two intrinsic, single-gate, nonlinear FET-models embedded in a network representing the device parasitics. A step-by-step procedure has been used to determine the 47 parameters of the model. DC-measurements were used to find starting values for some of the parameters of the nonlinear models. The parasitic capacitances were determined from three-port S-parameters measured at V_DS=0 V, I_DS=0 A and V(G₁S)=V(G₂S)=-4.0V. The parasitic inductances and resistances were determined from S-parameters measured at the same bias-point but with forward-biased gates, and from DC-measurements. The final model-optimization was done by simultaneously fitting the model to drain-to-source currents and three-port S-parameters measured at several different, active bias-points (V_DS >0)     

Analog performance of an ultrafast sampled-time all-optical fiber XPM wavelength converter

Analog performance of an all-optical ultrafast wavelength converter is measured and reported for the first time. The wavelength-conversion process is based on nonlinear cross-phase modulation in an optical fiber combined with an optical filter to convert phase modulation to amplitude modulation. The spurious-free dynamic range (SFDR) of the converter is measured to be 82 dB/spl middot/Hz/sup 2/3/. We define a new metric called the SFDR power penalty, which measures the degradation in SFDR relative to baseline the back-to-back analog optical link. The SFDR power penalty was measured to be 5 dB/spl middot/Hz/sup 2/3/ and is shown to be a function of the input optical power. This metric is used to characterize the linear region of the optical wavelength converter.     

大功率激光光束聚焦光斑功率密度分布直接测量仪的研究

针对激光加工大功率激光功率密度分布测量的要求 ,采用空心探针扫描采样测量法 ,提出并建立了被测激光经探针小孔、探针内通道传输的数学物理模型 ,经计算 ,给出了包括探针微孔孔径、探针内通道尺寸、系统采样点数等系统参数 ,设计了新的测量系统 ,实现了对大功率激光光束、聚焦光斑功率密度分布的直接测量 ,测量结果与理论计算相吻合。测量仪能对CO2 激光和YAG激光进行直接测量 ,测量的功率大于 10kW ,功率密度大于 10 7W /cm2 ,测量激光光束的最大直径为 6 0mm ,激光聚焦光斑的直径小于 0 5mm。     

Measurement of radiative and nonradiative recombination rate inInGaAsP-InP LED's

The curves of the minority carrier lifetime versus current are measured in InGaAsP-InP double heterostructure LEDs. To analyze the measured data the carrier recombination rate versus the minority carrier concentration n are calculated. Radiative and Auger recombination processes are considered to explain the measured data on heterointerfaces and/or on recombination centers. The radiative recombination rate R_rad versus n curves are confirmed to be parabolic. A simple analytical formula for the radiative recombination rate coefficient B(n) is derived     

Small-chirp 40-Gbps electroabsorption modulator with novel tensile-strained asymmetric quantum-well absorption layer

A small-chirp 40-Gbps electroabsorption modulator (EAM) with a novel tensile-strained asymmetric quantum-well (QW) absorption layer has been demonstrated for the first time. The strain and the band line-up of the asymmetric QW structure were designed in order to obtain a small-chirp operation, a clear eye opening, and a high extinction ratio simultaneously. The chirp measured as /spl alpha/-parameter was reduced without any penalty of extinction ratio and eye opening. The measured /spl alpha/-parameter was smaller than 1.5 at any bias voltage from 0 to -2 V. The measured 3-dB bandwidth of a 75-/spl mu/m-long EAM exceeded 50 GHz at -1 V bias voltage. Under a 40-Gbps modulation, a clear eye opening was obtained, and the eye diagram showed no violation of the standard STM256/OC768 mask. The measured dynamic extinction ratio was over 11 dB.     

Discontinuity Capacitance of a Coaxial Line Terminated in a Circular Waveguide

This calculation evaluates the discontinuity capacitance of a coaxial line terminated in a circular waveguide using the Rayleigh-Ritz variational technique. A 50-ohm 3/4-inch coaxial line termination with solid center conductor was fabricated with center and outer conductor dimensions of 0.32568/spl plusmn/ 0.00002 and 0.74995/spl plusmn/ 0.00002 inches, respectively. The measured value of capacitance of this termination at 1000 Hz was 2.164 x 10/sup -13/ farads as compared with the calculated value of 2.177093 x 10/sup -13/ farads. Calculated values of capacitance for other line sizes were also compared with measured values and in each case the calculated value agreed with the measured value to within the experimental error of the measured value.     

Packaged single-ended CMOS low noise amplifier with 2.3 dB noise figure and 64 dBm IIP/sub 2/

A single-ended low noise amplifier (LNA) implemented in a foundry 0.18 /spl mu/m CMOS process is tested on a PC board using the chip-on-board technique. The measured S/sub 11/ and S/sub 22/ are less than -10 dB over 5.15-5.35 GHz, which is the lower subband of UNII and HIPERLAN/2 band. The measured noise figure is 2.0 dB and power gain is 15.5 dB at 5.15 GHz, while drawing 5.8 mA of current from a 1.8 V supply. The measured IIP/sub 2/ is greater than 64 dBm. This extremely high IP/sub 2/ is due to the tuned response of the LNA. The LNA is suitable for WLAN applications in the lower UNII and HIPERLAN/2 subband.     

Low-noise In0.53Ga0.47As:Fe photoconductive detectors for optical communication

Photoconductive detectors were fabricated on In0.53- Ga0.47As/InP made high-resistive by doping with elemental Fe. A mobility of ∼ 6000 cm²/V . s and a net electron concentration of (2-5) × 10¹²cm^-3were measured in layers on which the devices were fabricated. Photoconductive gains of 5-10 were measured with CW and pulsed excitation. The calculation response time to 300-ps pulsed excitation is ∼ 100 ps and can be improved with reduced channel spacings. Typical dark currents in the devices are of the order of 50 µ A at room temperature. The noise power into a 50-Ω load measured at 82°C is -108.9 dBm. This is the lowest value measured in photoconductive detectors made with III-V In0.53Ga0.47As.     

Electromigration drift and threshold in Cu thin-film interconnects

The electromigration-induced ionic drift velocity and critical length-current density product, (jl_c) of Cu thin film conductors, were measured using the Blech-Kinsbron edge-displacement technique. Unencapsulated Cu edge-displacement segments on TiN conductors were stressed in vacuum at a modest current density of 6×10⁵ A/cm² in the temperature range of 175-275°C. Drift velocity was observed to be between 1-1/2 to 3 orders-of-magnitude lower than that previously measured for unencapsulated Al in this temperature range. We measured an activation energy for EM-induced drift of 1.25±0.08 eV which corresponds to grain boundary diffusion in Cu. Critical lengths were measured and the jl_c threshold was estimated to range between 900-1600 A/cm. We calculated a Cu grain boundary Z* value of -0.7, to our knowledge, this study is the first to measure Z* for electromigration in Cu thin film conductors     

Evaluation of the valence band discontinuity ofSi/Si1-xGex/Si heterostructures by application ofadmittance spectroscopy to MOS capacitors

In this study, admittance spectroscopy is applied for the first time to MOS capacitors fabricated on Si/Si_1-xGe_x/Si double-heterostructures, in order to evaluate the valence band discontinuity ΔE_v at the Si/Si_1-xGe_x interface. The principle of the measurement is presented and verified by the experimental results. A new feature of admittance spectroscopy applied to MOS capacitors is the ability to select the interface whose barrier is measured, by controlling the gate voltage. This fact is confirmed by the measurement of MOS capacitors, which include a SiGe well with different Ge contents at the front and the back interfaces. It is found from this measurement that, while ΔE_v at the back interface of the double-heterostructure is measured under slight depletion conditions for MOS capacitors, ΔE_v averaged between the front and the back interfaces is measured under accumulation conditions. The Ge_x content dependence of the measured ΔE_v is found to be in fairly good agreement with the theoretical values     

Polarity-dependent tunneling current and oxide breakdown indual-gate CMOSFETs

The gate tunneling leakage current in dual-gate CMOSFETs exhibits strong polarity dependence when measured in inversion, although it exhibits practically no polarity dependence when measured in accumulation. Specifically, p⁺-gate pMOSFET shows substantially lower tunneling current than n⁺-gate nMOSFET when measured in inversion. This polarity dependence arises from the difference in the supply of tunneling electrons. The polarity dependent tunneling current has a significant impact on oxide reliability measurements. For example, it gives rise to a higher T_bd value for p⁺/pMOSFET as compared to that for n⁺/nMOSFET when both are biased to inversion. Rationaless are given as to why T_bd is a better gauge than Q_bd for reliability assessment, and why nMOSFET is more prone to oxide breakdown than pMOSFET under normal operating conditions     

Linear and nonlinear optical properties of some ternary selenides

The refractive indices of the ternaryA^{I}B^{III}C_{2}^{VI}semiconductors CuAlSe2, AgGaSe2, CuGaSe2, and AgInSe2have been measured over most of the transparency range of these crystals. The optical nonlinear coefficients for second-harmonic generation of AgGaSe2CuGaSe2, and AgInSe2have also been measured. Three-frequency colinear phase matching is analyzed in detail for AgGaSe2. The birefringences of the other three crystals are not sufficient to permit three-frequency colinear phase matching within the range of the measured index. The merits of AgGaSe2for nonlinear optical applications are evaluated in comparison with other promising infrared nonlinear materials.     

半导体低维结构的压力光谱研究

研究了一些半导体低维结构的压力光谱．测得平均直径为26、52和62nm的In0.55Al0.45As／Al0.5Ga0.5As量子点发光峰的压力系数分别为82、94和98meV／GPa．表明这些发光峰具有Г谷的特性,这些量子点为Ⅰ型量子点．而平均直径为7nm的量子点发光峰的压力系数为-17meV／GPa,具有X谷的特性．所以这种小量子点为Ⅱ型量子点．测得ZnS：Mn纳米粒子中Mn发光峰的压力系数为-34．6meV／GPa,与晶体场理论的预计一致．而DA对发光峰基本不随压力变化,表明它应该与ZnS基体中的表面缺陷有关．测得ZnS：Cu纳米粒子中Cu的发光峰的压力系数为63．2meV／GPa,与ZnS体材料的带隙压力系数相同．表明Cu引入的受主能级具有浅受主的某些特点．测得ZnS：Eu纳米粒子中Eu发光峰的压力系数为24．1meV／GPa,与晶体场理论的预计不同．可能和Eu的激发态与ZnS导带间的相互作用有关．     

Cell Boundary Detection in the German Cellular Mobile Radio: System C

The efficiency of frequency utilization in mobile radio strongly depends on the performance of handoffs. Criteria for handoffs are usually derived from measured carrier-to-interference ratios (C/I). However, in mobile radio signal levels vary so rapidly that channel allocation cannot be performed in real time for optimum C/I. On the other hand, too many handoffs reduce frequency efficiency. The determination of cell boundaries, which are not only governed by C/I, but allocated to geographical lines in the terrain avoid unnecessary handoffs. In addition to C/I auxilary data are measured in the German system C and processed to detect the cell boundaries. The improvements gained from this procedure are demonstrated by measured results.     

Bias and temperature dependence of Sb-based heterostructure millimeter-wave detectors with improved sensitivity

Nearly lattice-matched InAs/AlSb/GaSb-based heterostructure backward diodes for zero-bias millimeter wave detection were fabricated and measured. A record-high curvature, /spl gamma/=39.1 V/sup -1/, at zero bias was measured. On-wafer sensitivity measurements from 1 to 110 GHz gave a record-high average sensitivity of 3687 V/W for zero-bias operation. Further enhancement of detector sensitivity was observed with applied dc bias, with a sensitivity of 7996 V/W obtained for a 0.9 /spl mu/A bias. Extrapolating the conjugately-matched measured sensitivity suggests that 1000 V/W should be achievable at a record-high 541 GHz. The temperature dependence of detector sensitivity was evaluated from measured dc current-voltage characteristics and gave expected sensitivities ranging from 3910 V/W at 293 K to 7740 V/W at 4.2 K.     

Gain, refractive index, and /spl alpha/-parameter in InGaAs-GaAs SQW broad-area lasers

Gain, refractive index, and the linewidth enhancement factor, or /spl alpha/-parameter, are measured in broad-area InGaAs-GaAs single-quantum-well semiconductor lasers using below-threshold amplified spontaneous emission spectra and a far-field filtering technique. The /spl alpha/ parameter is shown to increase dramatically with increasing carrier density and wavelength. Modes propagating in the transparent substrate of the lasers are shown to have a significant influence on the measured value of /spl alpha/.     

A GSM-GPRS/UMTS FDD-TDD/WLAN 802.11a-b-g multi-standard carrier generation system

A compact carrier generation system enabling proper interoperability among quad-band GSM, WCDMA (FDD and TDD), and WLAN (802.11a/b/g) standards is developed. The implementation is achieved in 0.25-/spl mu/m BiCMOS-SiGe process. The measured tuning range is higher that 1 GHz (3.05 to 4.1 GHz) exceeding the specifications by 25%. The voltage-controlled oscillator (VCO) exhibits a phase noise of -118 and -125 dBc/Hz measured, respectively, at 400 kHz and 1 MHz offsets while drawing 2.5 mA from 2.5 V supply. The measured phase noise at 400 kHz offset of the PCS/DCS output local-oscillator (LO) signal and the GSM output LO signal is, respectively, -124 dBc/Hz and -130 dBc/Hz.     

Experimental method to thermally deembed pads from R/sub TH/ measurements

Both compound semiconductor and silicon-based bipolar junction transistors or heterojunction bipolar transistors (HBTs) require the efficient removal of heat in order to achieve a maximum level of performance and reliability. In order to satisfy both of these criteria, the electrothermal behavior of each device must be captured in a compact model. The model parameter that determines the junction temperature is R/sub TH/, the thermal resistance. Experimental methods to determine R/sub TH/ often require a relatively small device with a large R/sub TH/ to be attached to a set of relatively large metal pads with a low R/sub TH/. The pads act as a thermal shunt to the substrate and artificially lower the measured R/sub TH/. In order to obtain a suitable R/sub TH/ value for a device located in an IC, the pads must be deembedded from the measured data, much like pad deembedding for an S-parameter measurement. Test structures with various width metal traces between the emitter pad and device's emitter have been fabricated in a 200-GHz InP double HBT process. A method of using the measured R/sub TH/ of these structures and a simple resistive network model to deembed the pads is presented. It is shown that deembedded values can be as much as 30% higher than the measured R/sub TH/.