U-grooved SIT CMOS technology with 3 fJ and 49 ps (7 mW, 350 fJ)operation

Static induction transistor (SIT) CMOS is analyzed by a circuit simulation method. According to the results, the propagation delay time of the SIT CMOS could be represented as the ratio of the load capacitance to the transconductance. The U-grooved structure plays an important role in the fabrication of MOS SIT with large transconductance and small parasitic capacitance. U-grooved SIT CMOS has been fabricated by anisotropic plasma etching, and its switching speed has been evaluated by a 31-stage ring oscillator. A minimum ρ-τ product of 3 fJ/gate has been obtained for a design rule of 1-μm channel length. A minimum propagation delay time of 49 ps/gate has also been obtained at a dissipation power of 7 mW/gate, which corresponds to a ρ-τ product of 350 fJ/gate     

Switching delay of a nonlatching Josephson gate evaluated from ringoscillator operation

The authors present the experimental results for the switching delay of a dc-biased nonlatching Josephson gate (a coupled-superconducting quantum interference device gate). The measurement is executed by the use of a ring oscillator (RO) method. A frequency-to-voltage converter is used to evaluate the oscillation frequency of the RO. The circuit is designed and fabricated using a 2.5-kA/cm² Nb/AlO_x/Nb Josephson junction technology. The results show the minimum switching delay of 18 ps     

Effects of gate geometry on propagation delay of integrated injection logic (I/SUP 2/L)

The effects of gate geometry on the propagation delay have been investigated for I/SUP 2/L gates with a self-aligned double-diffusion injector (S/SUP 2/L). To improve the switching speed of the I/SUP 2/L gate, the stored charge in the upside-down operated n-p-n transistor in the gate should be minimized. Following this principle, one can straightforwardly find that the reduction of the stored charges in the internal n-p-n base region and in the lateral p-n-p base region is the step to be taken for the further improvement of the speed. This can be realized by simply contracting the geometry of the gate. The minimum delay time realized in the gate was 3.2 ns/gate. Assuming that capabilities of processing the devices with 1-/spl mu/m accuracy become available, it is predicted that 1 ns/gate delay time can be realized with an improved S/SUP 2/L gate.     

New Technology Towards GaAs LSI/VLSI for Computer Applications

For future large-scale computer applications, new device technologies towards GaAs LSI/VLSI have been developed self-aligned fully implanted planar GaAs MESFET technology and high electron mobility transistor (HFMT) technology by molecular beam epitaxy (MBE). The self-aligned GaAs MESFET logic with 1.5-µm gate length exhibits a minimum switching time of 50 ps and the lowest power-delay product of 14.5 fJ at room temperature. The enhancement/depletion (E/D) type direct coupled HEMT logic has achieved a switching time of 17.1 ps with 1.7-µm gate length at liquid nitrogen temperature and more recently a switching time of 12.8 ps with 1.1-µm gate HEMT logic, which exceeds the top speed of Josephson Junction logic and shows the highest speed of any device logic ever reported. Optimized system performances are also projected to system delay of 200 ps at 10-kilogate integration with GaAs MESFET VLSI, and 100 ps at 100-kilogate with HEMT VLSI. These values of system delay correspond to the computer performance of over 100 million instructions per second (MIPS).     

New technology towards GaAs LSI/VLSI for computer applications

For future large-scale computer applications, new device technologies towards GaAs LSI/VLSI have been developed: self-aligned fully implanted planar GaAs MESFET technology and high electron mobility transistor (HEMT) technology by molecular beam epitaxy (MBE). The self-aligned GaAs MESFET logic with 1.5-µm gate length exhibits a minimum switching time of 50 ps and the lowest power-delay product of 14.5 fJ at room temperature. The enhancement/depletion (E/D) type direct coupled HEMT logic has achieved a switching time of 17.1 ps with 1.7-µm gate length at liquid nitrogen temperature and more recently a switching time of 12.8 ps with 1.1-µm gate HEMT logic, which exceeds the top speed of Josephson Junction logic and shows the highest speed of any device logic ever reported. Optimized system performances are also projected to system delay of 200 ps at 10-kilogate integration with GaAs MESFET VLSI, and 100 ps at 100-kilogate with HEMT VLSI. These values of system delay correspond to the computer performance of over 100 million instructions per second (MIPS).     

GaAs/AlGaAs and InGaAs/AlGaAs MODFET inverter simulations

Although MODFET's have exhibited the fastest switching speed for any digital circuit technology, there is as yet no clear consensus on optimal inverter design rules. We therefore have developed a comprehensive MODFET device model that accurately accounts for such high gate bias effects as transconductance degradation and increased gate capacitance. The device model, which agrees with experimental devices fabricated in this laboratory, is used in the simulation of direct-coupled FET logic (DCFL) inverters with saturated resistor loads. Based on simulation results, the importance of large driver threshold voltage not only for small propagation delay times but for wide logic swings and noise margins is demonstrated. Furthermore, minimum delay times are found to occur at small supply voltages as seen experimentally. Both of these results are attributed to the reduction of detrimental high gate bias effects. The major effect of reducing the gate length on delay time is to decrease the load capacitance of the gate. Using 0.25-µm gates, delay times of 5 and 3.6 ps at 300 and 77 K, respectively, are predicted. Finally, the recently introduced In-GaAs/AlGaAs MODFET's are shown to have switching speeds superior to those of conventional GaAs/AlGaAs MODFET's.     

Role of the amplifying gate in the turn-on process of involutestructure thyristors

The switching characteristics of involute thyristors with and without the amplifying gate structure are discussed. The effects of peak gate currents (10-100 A) on the anode current di/dt, switching delay, and energy loss in both types of devices are presented. The performance of the devices without the amplifying gate was far superior than that of the devices with the amplifying gate. A model is presented to explain this difference. Thyristors without the amplifying gate successfully switched anode currents on the order of 12.6 kA, at a di/dt of 100000 A/μs, from an anode voltage of 2 kV on a single-shot basis     

Fabrication technology for an 80-ps normally-off GaAs MESFET logic

Fabrication technology of a high-speed normally-off GaAs MESFET logic has been described. Anodic Oxidation process is applied to control epitaxial layer thickness precisely. A SiO2cap during alloying ohmic metal is used to prevent the ohmic layer surface from becoming uneven. A sloped mesa structure edge is used to avoid disconnection of metal interconnection. Electron-beam direct writing is employed to define a submicrometer gate. Applying these technologies, high-speed and small switching energy have been accomplished. The minimum delay timd and the associated switching energy were 77 ps and 75 fJ at room temperature and 51 ps and 97 fJ at 77 K.     

Analysis of highly doped collector transistors by usingtwo-dimensional process/device simulation and its application of ECLcircuits

A report is presented of the results of an investigation of device parameters and collector-to-emitter breakdown voltages of double polysilicon self-aligned transistors with highly doped collectors using a two-dimensional process/device simulation system. Favourable phosphorous-ion implanting condition for a highly doped pedestal collector was found to achieve a high cutoff frequency as well as low AC base resistance and small base-collector capacitance, thereby keeping the minimum collector-to-emitter breakdown voltage of 3 V. The authors also report ECL circuit performance improvements achieved in experiments that realized a minimum ECL gate delay time of 26.3 ps/gate at switching current of 1.64 mA as a result of process optimization. Moreover, a 1/8 static frequency divider T-F/F has been observed to operate up to a maximum frequency of 15.8 GHz     

体硅衬底上的CMOS Fin FET

介绍了一种制作在普通体硅上的 CMOS Fin FET.除了拥有和原来 SOI上 Fin FET类似的 Fin FET结构 ,器件本身在硅衬底中还存在一个凹槽平面 MOSFET,同时该器件结构与传统的 CMOS工艺完全相容 ,并应用了自对准硅化物工艺 .实验中制作了多种应用该结构的 CMOS单管以及 CMOS反相器、环振电路 ,并包括常规的多晶硅和 W/Ti N金属两种栅电极 .分析了实际栅长为 110 nm的硅基 CMOS Fin FET的驱动电流和亚阈值特性 .反相器能正常工作并且在 Vd=3V下 2 0 1级 CMOS环振的最小延迟为 14 6 ps/门 .研究结果表明在未来 VL SI制作中应用该结构的可行性     

体硅衬底上的CMOS FinFET

介绍了一种制作在普通体硅上的CMOS FinFET.除了拥有和原来SOI上FinFET类似的FinFET结构,器件本身在硅衬底中还存在一个凹槽平面MOSFET,同时该器件结构与传统的CMOS工艺完全相容,并应用了自对准硅化物工艺.实验中制作了多种应用该结构的CMOS单管以及CMOS反相器、环振电路,并包括常规的多晶硅和W/TiN金属两种栅电极.分析了实际栅长为110nm的硅基CMOS FinFET的驱动电流和亚阈值特性.反相器能正常工作并且在Vd=3V下201级CMOS环振的最小延迟为146ps/门.研究结果表明在未来VLSI制作中应用该结构的可行性.     

Dependence of normally-off GaAs JFET performance on device structure

The relation between the performance of normally-off JFET's and the Si ion-implantation conditions used to form the channel layer was studied. Static and switching characteristics were investigated for JFET's with three kinds of channel layers; Si implanted at 130 keV to doses of 2,4, and 6 × 10¹²ions/cm². While higher doses gave better static characteristics [Ids, gm, and Ron], higher capacitance degraded the switching characteristics. The optimum parameters were determined for the high-speed switching JFET. With 2-µm gate length, the highest switching speed was 80 ps and the lowest power-delay product was 0.9 fJ. An improved structure satisfying a high-conductance and low-capacitance requirement was successfully fabricated and showed excellent performance for high-speed and low-power logic circuits; the minimum propagation delay was 45 ps and the minimum power-delay product was 3.8 fJ with a delay time of 83 ps.     

A new model of switching operation in fully depleted ultrathin-filmCMOS/SIMOX

An analysis is made of the switching performances of fabricated ultrathin-film submicrometer-gate CMOS/SIMOX ring oscillators. A time-dependent gate capacitance model is proposed to explain the switching operation mechanism. It is found that reducing the gate capacitance by full depletion of the body silicon dramatically improves the propagation delay time of CMOS/SIMOX     

High speed I2L gate with self-aligned double-diffusion injector

To improve the switching speed of I2L gate, the charge storage in the n?p?n transistor should be minimised. For the experimental verification of a high-speed I2L gate, a very small I2L gate with p+ external base and heavily doped emitter regions was fabricated. The minimum propagation delay realised in the gate was 3.2 n s/gate.     

Some considerations on high-speed injection logic

After a brief review of the factors that limit the switching speed of standard I/SUP 2/L, the propagation delay time of some special high-speed I/SUP 2/L gates is computed. For a gate realized in oxide-isolated, shallow epitaxial layers, the delay time is directly dependent on the injector base width. Generally, the n-p-n switching transistor hardly contributes to the time delay. For a modified I/SUP 2/L gate in which saturation of the injector is avoided, the delay time is mainly determined by the unity gain frequency of the switching transistor. However, due to the heavy saturation of this transistor, values of /spl tau//SUB d/ already realized indicate that the speed improvement is less than an order of magnitude.     

Performance comparison of highly integrated circuits: Silicon NMOS versus gallium arsenide normally-off MESFET technology

This study presents a performance comparison between highly integrated circuits on gallium arsenide and on silicon realized with normally-off MESFET's and n-channel MOSFET's, respectively. As a basis, a standard cell structure is chosen in order to obtain a realistic capacitive loading. This cell is scaled down twice from an area of 1 mm²to 0.38 mm²and to 0.13 mm². The corresponding effective gate length inside the cell is 1, 0.5, and 0.2 µm, respectively. The delay time of a loaded inverter, the power consumption as well as the power-delay product are calculated using device parameters deduced from experimentalI-Vcharacteristics. For MOSFET's good noise margins at low switching times are obtained at a supply voltage of 3 V. The GaAs circuit exhibits a lower power consumption by one order of magnitude and a smaller delay time by about a factor of 2. Since nonoptimized GaAs MESFET's with recessed gates were regarded for the comparison improvements are expected for self-aligned MESFET's. For a supply voltage of 1 V, the MOSFET circuit shows a comparable power consumption to the GaAs circuit but longer delay time (factor 2 to 5).     

A 30 V Self-Aligned Metal/Poly-Si Replacement Gate Planar DMOS for DC/DC Converters

In this letter, a novel self-aligned metal/poly-Si gate planar double-diffused MOS (DMOS) is proposed and demonstrated for high-switching-speed and high-efficiency dc/dc converter applications. The self-aligned metal/poly-Si gate is realized by a replacement gate technology. The fabricated metal/poly-Si gate planar DMOS has a breakdown voltage of 36 V and a threshold voltage of 2.1 V. The gate sheet resistance of the metal/poly-Si gate is around 0.2 Omega/square, which is 50 times lower than that of the polysilicon gate. The low sheet resistance reduces the switching time as well as the power loss of the device during switching. For a device with a drain current of 69 A/cm², the turn-on and turn-off times are reduced from 29 to 25 ns and from 36 to 31 ns, respectively. The turn-on and turn-off switching energy losses are reduced by 22% and 15%, respectively     

集成电路时间延迟优化分析与模拟

基于Elmore模型，优化分析了N级二维CMOS传输门链和Ⅳ门三维双栅SOI IC的时间延迟，给出了HSPICE模拟结果。研完表明，由相同尺寸管子构成的N级二维CMOS门链，当把N级分作每3级为一组并且以缓冲门相间隔时，总时延存在极小值；由宽度尺寸比为3的三级不等尺寸管子所构造的传输门链间隔以缓冲门，也存在最小时延；当N门三维双栅SOI IC分为6个器件层时，可获得最小的时间延迟。     

Gate sizing using geometric programming

A two-step transistor sizing optimization method based on geometric programming for delay/area minimization is presented. In the first step, Elmore delay is minimized using only minimum and maximum transistor size constraints. In the second step, the minimized delay found in the previous step is used as a constraint for area minimization. In this way, our method can target simultaneously both delay and area reduction. Moreover, by relaxing the minimized delay, one may further reduce area with small delay penalty. Gate sizing may be accomplished through transistor sizing tying each transistor inside a cell to a same scale factor. This reduces the solution space, but also improves runtime as less variables are necessary. To analyze this tradeoff between execution time and solution quality a comparison between gate sizing and transistor sizing is presented. In order to qualify our approach, the ISCAS??85 benchmark circuits are mapped to a 45?nm technology using a typical standard cell library. Gate sizing and transistor sizing are performed considering delay minimization. Gate sizing is able to reduce delay in 21?%, on average, for the same area and power values of the sizing provided by standard-cells library. Then, the transistor sizing is executed and delay can be reduced in 40.4?% and power consumption in 2.9?%, on average, compared to gate sizing. However, the transistor sizing takes about 23 times longer to be computed, on average, using a number of variables twice higher than gate sizing. Gate sizing optimizing area is executed considering a delay constraint. Three delay constraints are considered, the minimum delay given by delay optimization and delay 1 and 5?% higher than minimum delay. An energy/delay gain (EDG) metric is used to quantify the most efficient tradeoff. Considering the minimum delay, area (power) is reduced in 28.2?%, on average. Relaxing delay by just 1?%, area (power) is reduced in 41.7?% and the EDG metric is 41.7. Area can be reduced in 51?%, on average, relaxing delay by 5?% and EDG metric is 10.2.     

3.21 ps ECL gate using InP/InGaAs DHBT technology

A new circuit configuration for an emitter-coupled logic (ECL) gate that can reduce propagation delay time has been demonstrated. Nineteen-stage ring oscillators were fabricated using InP/InGaAs double-heterojunction bipolar transistors (DHBTs) with cutoff frequency f/sub T/ and maximum oscillation frequency f/sub max/ of about 232 and 360 GHz, respectively, to evaluate the speed performance of the proposed ECL gate. The minimum propagation delay is 3.21 ps/gate. The proposed ECL gate is about 8% faster than the conventional ECL gate.