Zeitschriftenaufsätze

For four decades semiconductor electronics has followed Moore' s law: with each generation of integration the circuit features became smaller, more complex and faster. This development is now reaching a wall so that smaller is no longer any faster. The clock rate has saturated at about 3-5 GHz and the parallel processor approach will soon reach its limit. The prime reason for the limitation the semiconductor electronics experiences are not the switching speed of the individual transistor, but its power dissipation and thus heat. Digital superconductive electronics is a circuit- and device-technology that is inherently faster at much less power dissipation than semiconductor electronics. It makes use of superconductors and Josephson junctions as circuit elements, which can provide extremely fast digital devices in a frequency range - dependent on the material - of hundreds of GHz: for example a flip-flop has been demonstrated that operated at 750 GHz. This digital technique is scalable and follows similar design rules as semiconductor devices. Its very low power dissipation of only 0.1 µWper gates at 100 GHz opens the possibility of three-dimensional integration. Circuits like icroprocessors and analogue-to-digital converters for commercial and military applications have been demonstrated. In contrast to semiconductor circuits, the operation of superconducting circuits is based on naturally standardized digital pulses the area of which is exactly the flux quantum 0. The flux quantum is also the natural quantization unit for digital-to-analogue and analogue-to-digital converters. The latter application is so precise, that it is being used as voltage standard and that the physical unit 'Volt' is defined by means of this standard. Apart from its outstanding features for digital electronics, superconductive electronics provides also the most sensitive sensor for magnetic fields: the Superconducting Quantum Interference Device (SQUID). Amongst many other applications SQUIDs are used as sensors for magnetic heart and brain signals in medical applications, as sensor for geological surveying and food-processing and for non-destructive testing. As amplifiers of electrical signals, SQUIDs can nearly reach the theoretical limit given by Quantum Mechanics. A further important field of application is the detection of very weak signals by 'transition-edge' bolometer, superconducting nanowire single-photon detectors, and superconductive tunnel junctions. Their application as radiation detectors in a wide frequency range, from microwaves to X-rays is now standard. The very low losses of superconductors have led to commercial microwave filter designs that are now widely used in the USA in base stations for cellular phones and in military communication applications. The number of demonstrated applications is continuously increasing and there is no area in professional electronics, in which superconductive electronics cannot be applied and surpasses the performance of classical devices. Superconductive electronics has to be cooled to very low temperatures. Whereas this was a bottleneck in the past, cooling techniques have made a huge step forward in recent years: very compact systems with high reliability and a wide range of cooling power are available commercially, from micro coolers of match-box size with milli-Watt cooling power to high-reliability coolers of many Watts of cooling power for satellite applications. Superconductive electronics will not replace semiconductor electronics and similar room-temperature techniques in standard applications, but for those applications which require very high speed, low-power consumption, extreme sensitivity or extremely high precision, superconductive electronics is superior to all other available techniques. To strengthen the European competitiveness in superconductor electronics research projects have to be set-up in the following field: - Ultra-sensitive sensing and imaging. - Quantum measurement instrumentation. - Advanced analogue-to-digital converters. - Superconductive electronics technology.

We investigated the gray zone width of Josephson comparators by means of circuit simulations and experiments, looking at the dependences on different circuit parameters and topologies. Eight different comparator circuits were simulated and designed for a 1 kA cm 2 niobium device. With our sophisticated measurement set-up, the lowest reported gray zone width of 3.2 A at 4.2 K was measurable. Moreover, the results obtained allow us to derive a set of design rules for further reduction of the gray zone width, which was the original goal of our investigations. (Some figures in this article are in colour only in the electronic version)