This paper describes a new soft-error-immune SICOS upward transistor memory cell suitable for ultra-high-speed bipolar RAMs. A cell size of 180 µm2, significantly smaller than the 500 µm2 in the conventional upward transistor cell, is achieved by marging an upward transistor and a Shottky barrier diode. A new very thin polysilicon resistor and 0.5-µm U-groove isolated SICOS technology are used to furher reduce cell size. The memory cell is about 105 times as immune to soft errors as downward transistor cells. A simulation shows that a 256-Kbit RAM with a write cycle time below 3 ns can be made using this memory cell.

Recent high-speed bipolar technologies based on SICOS (Sidewall Base Contact Structure) transistors are reviewed. Bipolar device structures that include polysilicon are key technologies for improving circuit characteristics. As the characteristics of the upward operated SICOS transistors are close to those of downward transistors, they can easily be applied in memory cells which have near-perfect soft-error-immunity. Newly developed process technologies for making shallow base and emitter junctions to improve circuit performance are also reviewed. Finally, complementary bipolar technology for low-power and high-speed circuits using pnp transistors, and a quasi-drift base transistor structure suitable for below 0.1 µm emitters are discussed.

We describe the design of ECL write circuits and a CMOS memory cell in an ECL-CMOS SRAM to achieve ultra-fast cycle time. Factors determining the write cycle are reduced by several novel circuit techniques and by optimizing the design of the write circuits and CMOS memory cell, thereby, enabling ultra-fast cycle time. Key techniques are a bit line overdriving, the use of an overshoot suppressing emitter follower and a WPG with a replica memory cell delayer. The 72-kb ECL-CMOS SRAM macro through which these techniques were implemented was fabricated using 0. 3-µm BiCMOS technology. The RAM macro achieves a short cycle time of 2 ns without sacrificing stable memory cell operation. These techniques thus provide SRAMs with a shorter cycle time in the cache memories of high performance computer systems.

Two high-speed sensing techniques suitable for ultrahigh-speed SRAM's are proposed. These techniques can reduce a 64-kb SRAM access time to 71 89% of that of conventional high-speed bipolar SRAM's. The techniques use a small CMOS memory cell instead of the bipolar memory cell that has often been used in conventional bipolar SRAM's for cache and control memories of mainframe computers. Therefore, the memory cell size can also be reduced to 26 43% of conventional cells. A 64-kb SRAM with one of the sensing techniques is fabricated using 0.5-µm BiCMOS technology and achieves a 1.5-ns access time with a 78-µm2 memory cell size. The techniques are especially useful in the development of both ultrahigh-speed and high-density SRAM's, which have been used as cache and control memories of mainframe computers.

A novel redundancy method suitable for an ultra-high-speed SRAM with logic gates is proposed. Fuse decoders are used to reduce the number of fuses, thus suppressing the access time degradation. This makes it possible to flip chip bond an SRAM with logic gates, which has a high pin count and operates at a very high frequency. To combine the new redundancy method and an ECL decoder circuit with a BiCMOS inverter, several schemes for disabling a defective cell and enabling a spare one are discussed. A 1-Mb ECL-CMOS SRAM with 120-k logic gates was fabricated using 0.3-µm BiCMOS technology. This SRAM consists of 16 RAM macros, and the RAM macro had an access time of only 0.65 ns. The access time degradation after repair was less than 50 ps.

A new redundancy technique especially suitable for ultra-high-speed static RAMs (SRAMs) has been developed. This technique is based on a decoding-method that uses two kinds of fuses without introducing any additional delay time. One fuse is initially ON and can be turned OFF afterwards, if necessary, by a cutting process using a focused ion beam (FIB). The other is initially OFF and can be turned ON afterwards by a connecting process using laser chemical vapor deposition (L-CVD). This technique is applied to a 64 kbit SRAM having a 1.5-ns access time. The experimental results obtained through an SRAM chip repaired using this redundancy technique show that this technique does not introduce any increase in the access time and does not reduce the operational margin of the SRAM.

An ultrahigh-speed 72-kb ECL-CMOS RAM macro for a 1-Mb SRAM with 0.65-ns address-access time, 0.80-ns write-pulse width, and 30.24-µm2 memory cells has been developed using 0.3-µm BiCMOS technology. Two key techniques for achieving ultrahigh speed are an ECL decoder/driver circuit with a BiCMOS inverter and a write-pulse generator with a replica memory cell. These circuit techniques can reduce access time and write-pulse width of the 72-kb RAM macro to 71% and 58% of those of RAM macros with conventional circuits. In order to reduce crosstalk noise for CMOS memory-cell arrays driven at extremely high speeds, a twisted bit-line structure with a normally on MOS equalizer is proposed. These techniques are especially useful for realizing ultrahigh-speed, high-density SRAM's, which have been used as cache and control storages in mainframe computers.

An ECL-CMOS SRAM technology is proposed which features a combination of ECL word drivers, ECL write circuits and low-voltage CMOS cells. This technology assures both ultra-high-speed and high-density. In the ECL-CMOS SRAM,various kinds of noise generated during the write cycle seriously affect the memory performance, because it has much faster access than conventional SRAMs. To overcome this problem, we propose three noise reduction techniques; a noise reduction clamp circuit, an emitter follower with damping capacitor and a twisted bit line structure with "normally on" equalizer. These techniques allow fast accese and cycle times. To evaluate these techniques, a 64-kb SRAM chip was fabricated using 0.5-µm BiCMOS technology. This SRAM has a short cycle time of 2 ns and a very fast access time of 1.5 ns. Evaluation proves the usefulness of these techniques.

This paper describes a collected charge enhancement due to the alpha-particle-induced charge transfer through an extremely thin (less than 0.2µm) p-layer sandwiched between two n-layers. This charge enhancement is caused by a parasitic transistor composed of the p- and the two n-layers which is turned on by the alpha-particle-induced charge. The collected charge enhancement occurs depending on the impurity concentration and the thickness of the thin layer. The condition whether the charge enhancement occurs or not is determined quantitatively using a 3-D device simulator. The simulation and experimental results show that, if the collected charge enhancement occurs, the soft-error rate is dominantly determined by it and cannot be decreased by increasing the cell stored charge.

BiCMOS circuits using a base-boost technique for low-voltage application have been proposed. These circuits can operate at supply voltages down to 1.5 V. Their power dissipation, however, is 1.5-2 times of that of the CMOS circuit. We propose a novel BiCMOS circuit dissipating less power than that of conventional circuits. A base-boost technique is a key to low-voltage operation, and a gate holding the output voltage and a depletion nMOS pre-charge transistor are also introduced to reduce the power dissipation. Results of simulations using 0.3µm BiCMOS device parameters show that base-boosted BiNMOS (BB-BiNMOS) circuit is 1.5 times faster than CMOS circuit even at 1 V and that its power dissipation is almost the same power as that of a CMOS circuit, the base-boosted BiCMOS (BB-BiCMOS) circuit is twice as fast and dissipates only 1.2 times as much power. The energy-delay product of the BB-BiCMOS circuit is smaller than that of conventional BiCMOS circuits and is about half of that of a CMOS circuit, the BB-BiCMOS circuit is thus the most promising high-speed circuits for low-voltage and low-power applications.