Techniques to Extend Canary-Based Standby $V_{DD}$ Scaling for SRAMs to 45 nm and Beyond

$V_{DD}$ scaling is an efficient technique to reduce SRAM leakage power during standby mode. The data retention voltage (DRV) defines the minimum $V_{DD}$ that can be applied to an SRAM cell without losing data. The conventional worst-case guard-banding approach selects a fixed standby supply voltage at design time to accommodate the variability of DRV, which sacrifices potential power savings for non-worst-case scenarios. We have proposed a canary-based feedback to achieve aggressive power savings by tracking PVT variations through canary cell failures. In this paper, we show new measured silicon results that confirm the ability of the canary scheme to track PVT changes. We thoroughly analyze the adaptiveness of the canary cells for tracking changes in the SRAM array, including the ability to track PVT fluctuations. We present circuits for robustly building the control logic that implements the feedback mechanism at subthreshold supply voltages, and we derive a new analytical model to help tune the canary cells in the presence of variations. To realistically quantify the potential savings achievable by the canary scheme, we assess the impact of various sources of overhead. Finally, we investigate the performance of the canary based scheme in nanometer technologies, and we show that it promises to provide substantial standby power savings down to the 22 nm node.      

Low-voltage high-gain differential OTA for SC circuits

A new differential operational transconductance amplifier (OTA) for SC circuits that operates with a supply voltage of less than two transistor threshold voltages is presented. Its simplicity relies on the use of a low-voltage regulated cascode circuit, which achieves very high output impedance under low-voltage restrictions. The OTA has been designed to operate with a supply voltage of V/sub DD/=1.5 V, using a 0.6 /spl mu/m CMOS technology with transistor threshold voltages of V/sub TN/=0.75 V and V/sub TP/=-0.85 V. Post-layout simulation results for a load capacitance (C/sub L/) of 2 pF show a 75 MHz gain-bandwidth product and 100 dB DC gain with a quiescent power consumption of 750 /spl mu/W.     

Low-power low-voltage differential class-AB OTAs for SC circuits

Two new differential class-AB operational transconductance amplifiers (OTAs) for SC circuits that operate with a supply voltage of less than two transistor threshold voltages are introduced. They make use of a new class-AB pseudodifferential pair to generate signal currents much larger than quiescent currents. Both OTAs have been designed to operate with a supply voltage of V/sub DD/=1.1 V, using a 0.35 /spl mu/m CMOS technology. Simulation results for a load capacitance (C/sub L/) of 1 pF show 15 MHz gain-bandwidth product with a quiescent power consumption of 10 /spl mu/W.     

A 0.5-V 1-/spl mu/W successive approximation ADC

A successive approximation analog-to-digital converter (ADC) is presented operating at ultralow supply voltages. The circuit is realized in a 0.18-/spl mu/m standard CMOS technology. Neither low-V/sub T/ devices nor voltage boosting techniques are used. All voltage levels are between supply voltage V/sub DD/ and ground V/sub SS/. A passive sample-and-hold stage and a capacitor-based digital-to-analog converter are used to avoid application of operational amplifiers, since opamp operation requires higher values for the lowest possible supply voltage. The ADC has signal-to-noise-and-distortion ratios of 51.2 and 43.3 dB for supply voltages of 1 and 0.5 V, at sampling rates of 150 and 4.1 kS/s and power consumptions of 30 and 0.85 /spl mu/W, respectively. Proper operation is achieved down to a supply voltage of 0.4 V.     

Two-phase boosted voltage generator for low-voltage DRAMs

A two-phase boosted voltage (V/sub PP/) generator circuit was proposed for use in gigabit DRAMs. It reduced the maximum gate-oxide voltage of pass transistor and the lower limit of supply voltage to V/sub PP/ and V/sub TN/, respectively, while those for the conventional charge-pump circuit are V/sub PP/+V/sub DD/ and 1.5 V/sub TN/ respectively. Also, the pumping current was increased in the new circuit. The newly proposed two-phase V/sub PP/ charge-pump circuit worked successfully at V/sub DD/ down to 0.8 V by eliminating the threshold voltage loss of the control pulse generator and was tested successfully in a 0.16-/spl mu/m test chip using triple-well CMOS technology.     

Modeling and sizing for minimum energy operation in subthreshold circuits

This paper examines energy minimization for circuits operating in the subthreshold region. Subthreshold operation is emerging as an energy-saving approach to many energy-constrained applications where processor speed is less important. In this paper, we solve equations for total energy to provide an analytical solution for the optimum V/sub DD/ and V/sub T/ to minimize energy for a given frequency in subthreshold operation. We show the dependence of the optimum V/sub DD/ for a given technology on design characteristics and operating conditions. This paper also examines the effect of sizing on energy consumption for subthreshold circuits. We show that minimum sized devices are theoretically optimal for reducing energy. A fabricated 0.18-/spl mu/m test chip is used to compare normal sizing and sizing to minimize operational V/sub DD/ and to verify the energy models. Measurements show that existing standard cell libraries offer a good solution for minimizing energy in subthreshold circuits.     

An offset compensation technique for latch type sense amplifiers in high-speed low-power SRAMs

The input referred offset voltage occurring in the full latch V/sub DD/ biased sense amplifier has been analyzed extensively. The process variations in the matched nMOS and pMOS transistors have been accounted by /spl plusmn/2.5% variation in V/sub T/ and /spl plusmn/5% variation in /spl beta/, from typical values. Effect of various design parameters on the sense amplifier offset has been studied and reported. It has been shown that the rise time of the sense amplifier enable signal (SAEN) has a profound effect on the offset voltage. The slower transition of SAEN signal is proposed to result in high speed as well as low-power consumption in SRAM application. An analytical model has been derived for simplified latch to model the effect of rise time of SAEN signal on offset voltage.     

A low-power SRAM using hierarchical bit line and local sense amplifiers

This paper proposes a low power SRAM using hierarchical bit line and local sense amplifiers (HBLSA-SRAM). It reduces both capacitance and write swing voltage of bit lines by using the hierarchical bit line composed of a bit line and sub-bit lines with local sense amplifiers. The HBLSA-SRAM reduces the write power consumption in bit lines without noise margin degradation by applying a low swing signal to the high capacitive bit line and by applying a full swing signal to the low capacitive sub-bit line. The HBLSA-SRAM reduces the swing voltage of bit lines to V/sub DD//10 for both read and write. It saves 34% of the write power compared to the conventional SRAM. An SRAM chip with 8 K/spl times/32 bits is fabricated in a 0.25-/spl mu/m CMOS process. It consumes 26 mW read power and 28 mW write power at 200 MHz with 2.5 V.     

High breakdown voltage (Al/sub 0.3/Ga/sub 0.7/)/sub 0.5/In/sub 0.5/P/InGaAs quasi-enhancement-mode pHEMT with field-plate technology

A high breakdown voltage and a high turn-on voltage (Al/sub 0.3/Ga/sub 0.7/)/sub 0.5/In/sub 0.5/P/InGaAs quasi-enhancement-mode (E-mode) pseudomorphic HEMT (pHEMTs) with field-plate (FP) process is reported for the first time. Between gate and drain terminal, the transistor has a FP metal of 1 /spl mu/m, which is connected to a source terminal. The fabricated 0.5/spl times/150 /spl mu/m/sup 2/ device can be operated with gate voltage up to 1.6 V owing to its high Schottky turn-on voltage (V/sub ON/=0.85 V), which corresponds to a high drain-to-source current (I/sub ds/) of 420 mA/mm when drain-to-source voltage (V/sub ds/) is 3.5 V. By adopting the FP technology and large barrier height (Al/sub 0.3/Ga/sub 0.7/)/sub 0.5/In/sub 0.5/P layer design, the device achieved a high breakdown voltage of -47 V. The measured maximum transconductance, current gain cutoff frequency and maximum oscillation frequency are 370 mS/mm, 22 GHz , and 85 GHz, respectively. Under 5.2-GHz operation, a 15.2 dBm (220 mW/mm) and a 17.8 dBm (405 mW/mm) saturated output power can be achieved when drain voltage are 3.5 and 20 V. These characteristics demonstrate that the field-plated (Al/sub 0.3/Ga/sub 0.7/)/sub 0.5/In/sub 0.5/P E-mode pHEMTs have great potential for microwave power device applications.     

1677 V, 5.7 m/spl Omega//spl middot/cm/sup 2/ 4H-SiC BJTs

This letter reports the development of a high-performance power 4H-SiC bipolar junction transistor (BJT) with, simultaneously, a high blocking voltage and a low specific on-resistance (R/spl I.bar//sub ON/). A single BJT cell with an active area of 0.61 mm/sup 2/ achieves an open base collector-to-emitter blocking voltage (V/sub ceo/) of 1677 V and conducts up to 3.2 A at a forward voltage drop of V/sub CE/=3.0 V, corresponding to a low R/spl I.bar//sub ON/ of 5.7 m/spl Omega//spl middot/cm/sup 2/ up to Jc=525 A/cm/sup 2/ and a record high value of V/sub B//sup 2//R/sub SP/spl I.bar/ON/ of 493 MW/cm/sup 2/.     

Mixed body bias techniques with fixed V/sub t/ and I/sub ds/ generation circuits

There remains a need to improve sub-1-V CMOS VLSIs with respect to variation in transistor behavior. In this paper, to minimize variation in delay and the noise margin of the circuits in processors, we propose several mixed body bias techniques using body bias generation circuits. In these circuits, either the saturation region of the current between source and drain (I/sub ds/) or the threshold voltage (V/sub t/) of PMOS/NMOS is permanently fixed, regardless of temperature range or variation in process. A test chip that featured these body bias generation circuits was fabricated using a 130-nm CMOS process with a triple-well structure. The mixed body bias techniques which keep the I/sub ds/ of the MOS in the decoder and I/O circuits of a register file fixed and maintain the V/sub t/ of the MOS in both the memory cell and domino circuits of the register file fixed resulted in positive temperature dependence of delay from -40 /spl deg/C to 125 /spl deg/C, 85% reduction of the delay variation compared with normal body bias (NBB) at V/sub DD/ = 0.8 V. In addition, the results using these techniques show a 100-mV improvement in lower operating voltage compared with NBB at -40 /spl deg/C on a 4-kb SRAM.     

Characterization of CMOS Devices for VLSI

CMOS bulk and SOS technologies are discussed for VLSI with emphasis on static and dynamic characteristics of two-input NAND gates. Olpthnum performance (minimum figure of merit FM= f/sub pd/P/sub d/) is obtained for a CMOS/SOS two-input NAND gate (FO = 2, C/sub L/ = 22 fF) with an electrical channel length L = 0.75 /spl mu/m, channel width W= 5.0 /spl mu/m, and oxide thickness X/sub O/ = 450 /spl Aring/with V/sub DD/ = 3.0 V, to yield t/sub pd/ = 400 ps and P/sub d/ = 250 /spl mu/W (t/sub pd/P/sub d/ = 100 fJ) at room temperature. Bulk technology performs within a factor of 2 of SOS for t/sub pd/ and P/sub d/. CMOS technologies offer subnanosecond propagation delays, similar to ECL bipolar, at the low submilliwatt power levels of CMOS. An analytical expression for t/sub pd/ describes the performance of two-input NAND gates in terms of device modeling and fabrication parameters. Such an expression provides a hierarchal modeling approach to characterize mini-cells for VLSI.     

Electrical characteristics of 8-/spl Aring/ EOT HfO/sub 2//TaN low thermal-budget n-channel FETs with solid-phase epitaxially regrown junctions

The authors demonstrate high-performing n-channel transistors with a HfO/sub 2//TaN gate stack and a low thermal-budget process using solid-phase epitaxial regrowth of the source and drain junctions. The thinnest devices have an equivalent oxide thickness (EOT) of 8 /spl Aring/, a leakage current of 1.5 A/cm/sup 2/ at V/sub G/=1 V, a peak mobility of 190 cm/sup 2//V/spl middot/s, and a drive-current of 815 /spl mu/A//spl mu/m at an off-state current of 0.1 /spl mu/A//spl mu/m for V/sub DD/=1.2 V. Identical gate stacks processed with a 1000-/spl deg/C spike anneal have a higher peak mobility at 275 cm/sup 2//V/spl middot/s, but a 5-/spl Aring/ higher EOT and a reduced drive current at 610 /spl mu/A//spl mu/m. The observed performance improvement for the low thermal-budget devices is shown to be mostly related to the lower EOT. The time-to-breakdown measurements indicate a maximum operating voltage of 1.6 V (1.2 V at 125 /spl deg/C) for a ten-year lifetime, whereas positive-bias temperature-instability measurements indicate a sufficient lifetime for operating voltages below 0.75 V.     

Drain-bias dependence of threshold voltage stability of amorphous silicon TFTs

Amorphous silicon (a-Si:H) thin-film transistors (TFTs) used in emerging, nonswitch applications such as analog amplifiers or active loads, often have a bias at the drain terminal in addition to the gate that can alter their threshold voltage (V/sub T/) stability performance. At small gate stress voltages (0/spl les/V/sub ST//spl les/15 V) where the defect state creation instability mechanism is dominant, the presence of a bias at the TFT drain decreases the overall shift in V/sub T/(/spl Delta/V/sub T/) compared to the /spl Delta/V/sub T/ in the absence of a drain bias. The measured shift in V/sub T/ appears to agree with the defect pool model that the /spl Delta/V/sub T/ is proportional to the number of induced carriers in the a-Si:H channel.     

4H-SiC power bipolar junction transistor with a very low specific ON-resistance of 2.9 m/spl Omega//spl middot/cm/sup 2/

This letter reports a newly achieved best result on the specific ON-resistance (R/sub SP/spl I.bar/ON/) of power 4H-SiC bipolar junction transistors (BJTs). A 4H-SiC BJT based on a 12-/spl mu/m drift layer shows a record-low specific-ON resistance of only 2.9 m/spl Omega//spl middot/cm/sup 2/, with an open-base collector-to-emitter blocking voltage (V/sub ceo/) of 757 V, and a current gain of 18.8. The active area of this 4H-SiC BJT is 0.61 mm/sup 2/, and it has a fully interdigitated design. This high-performance 4H-SiC BJT conducts up to 5.24 A at a forward voltage drop of V/sub CE/=2.5 V, corresponding to a low R/sub SP-ON/ of 2.9 m/spl Omega//spl middot/cm/sup 2/ up to J/sub c/=859 A/cm/sup 2/. This is the lowest specific ON-resistance ever reported for high-power 4H-SiC BJTs.     

A technique to improve the drive current of high-voltage I/O transistors in digital CMOS technologies

By combining a 0.12-/spl mu/m-long 1.2-V thin-oxide transistor with a 0.22-/spl mu/m-long 3.3-V thick-oxide transistor in a 0.13-/spl mu/m CMOS process, a composite MOS transistor structure with a drawn gate length of 0.34 /spl mu/m is realized. Measurements show that at V/sub GS/=1.2 V and V/sub DS/=3.3 V, the composite transistor has more than two times the drain current of the minimum channel length (0.34 /spl mu/m) 3.3-V thick-oxide transistor, while having the same breakdown voltage (V/sub BK/) as the thick-oxide transistor. Exploiting these, it should be possible to implement 3.3-V I/O transistors with better combination of drive current, threshold voltage (V/sub T/) and breakdown voltage in conventional CMOS technologies without adding any process modifications.     

Delay and power monitoring schemes for minimizing power consumption by means of supply and threshold voltage control in active and standby modes

This paper describes newly developed delay and power monitoring schemes for minimizing power consumption by means of the dynamic control of supply voltage V/sub DD/ and threshold voltage V/sub TH/ in active and standby modes. In the active mode, on the basis of delay monitoring results, either VDD control or VTH control is selected to avoid any oscillation problem between them. In V/sub DD/ control, on the basis of delay monitoring results, VDD is adjusted so as to be maintained at the minimum value at which the chip is able to operate for a given clock frequency. In V/sub TH/ control, on the basis of power monitoring results, VTH is adjusted so as to maintain a certain switching current I/sub SW//leakage current I/sub LEAK/ ratio known to indicate minimum power consumption. In the standby mode, the precision of power monitoring (which detects optimum body bias by comparing subthreshold current I/sub SUBTH/ to substrate current I/sub SUB/) is improved by taking into consideration both the effects of lowering V/sub DD/ and the effects of the presence of gate-oxide leakage current. Experimental results with a 90-nm CMOS device indicate that use of the proposed power monitoring results in the successful minimizing of power consumption. It does so by making it possible to: 1) maintain the I/sub SW//ILEAK ratio in the active mode and 2) detect optimum body bias conditions (I/sub SUBTH/=ISUB) within an error of less than 20% with respect to actual minimum leakage current values in the standby mode.     

Latchup-free ESD protection design with complementary substrate-triggered SCR devices

The turn-on mechanism of silicon-controlled rectifier (SCR) devices is essentially a current triggering event. While a current is applied to the base or substrate of an SCR device, it can be quickly triggered on into its latching state. In this paper, latchup-free electrostatic discharge (ESD) protection circuits, which are combined with the substrate-triggered technique and an SCR device, are proposed. A complementary circuit style with the substrate-triggered SCR device is designed to discharge both the pad-to-V/sub SS/ and pad-to-V/sub DD/ ESD stresses. The novel complementary substrate-triggered SCR devices have the advantages of controllable switching voltage, adjustable holding voltage, faster turn-on speed, and compatible to general CMOS process without extra process modification such as the silicide-blocking mask and ESD implantation. The total holding voltage of the substrate-triggered SCR device can be linearly increased by adding the stacked diode string to avoid the transient-induced latchup issue in the ESD protection circuits. The on-chip ESD protection circuits designed with the proposed complementary substrate-triggered SCR devices and stacked diode string for the input/output pad and power pad have been successfully verified in a 0.25-/spl mu/m salicided CMOS process with the human body model (machine model) ESD level of /spl sim/7.25 kV (500 V) in a small layout area.     

A 5.7-GHz 0.18-/spl mu/m CMOS gain-controlled differential LNA with current reuse for WLAN receiver

This letter presents a 5.7 GHz 0.18 /spl mu/m CMOS gain-controlled differential LNA for an IEEE 802.11a WLAN application. The differential LNA, fabricated with the 0.18 /spl mu/m 1P6M standard CMOS process, uses a current-reuse technology to increase linear gain and save power consumption. The circuit measurement is performed using an FR-4 PCB test fixture. The LNA exhibits a noise figure of 3.7 dB, linear gain of 12.5 dB, P/sub 1dB/ of -11 dBm, and gain tuning range of 6.9 dB. The power consumption is 14.4 mW at V/sub DD/=1.8 V.     

Characterization of 1.9- and 1.4-nm ultrathin gate oxynitride by oxidation of nitrogen-implanted silicon substrate

For gate oxide thinned down to 1.9 and 1.4 nm, conventional methods of incorporating nitrogen (N) in the gate oxide might become insufficient in stopping boron penetration and obtaining lower tunneling leakage. In this paper, oxynitride gate dielectric grown by oxidation of N-implanted silicon substrate has been studied. The characteristics of ultrathin gate oxynitride with equivalent oxide thickness (EOT) of 1.9 and 1.4 nm grown by this method were analyzed with MOS capacitors under the accumulation conditions and compared with pure gate oxide and gate oxide nitrided by N/sub 2/O annealing. EOT of 1.9- and 1.4-nm oxynitride gate dielectrics grown by this method have strong boron penetration resistance, and reduce gate tunneling leakage current remarkably. High-performance 36-nm gate length CMOS devices and CMOS 32 frequency dividers embedded with 57-stage/201-stage CMOS ring oscillator, respectively, have been fabricated successfully, where the EOT of gate oxynitride grown by this method is 1.4 nm. At power supply voltage V/sub DD/ of 1.5 V drive current Ion of 802 /spl mu/A//spl mu/m for NMOS and -487 /spl mu/A//spl mu/m for PMOS are achieved at off-state leakage I/sub off/ of 3.5 nA//spl mu/m for NMOS and -3.0 nA//spl mu/m for PMOS.