High-Temperature Superconductors—Then and Now
Eric Bogatin, Contributing Editor -- Semiconductor International, 6/1/2001
 |
Since the discovery in 1911 by Heike Kamerlingh Onnes of mercury's superconductivity at ~4 K, the highest superconducting transition temperature has been inching up. It took 30 years for the temperature to get to 15 K with a niobium alloy. This was comfortably in the liquid helium range and allowed practical applications outside the research lab. After another 30 years, the highest transition temperature of any metal was still only 23 K.
Alex Muller and Georg Bednorz of IBM, with a new class of superconducting materials — ceramic metal oxides — shattered this record in 1986. Their sample of lanthanum, barium copper oxide transitioned at 30 K. Within a year, Paul Chu and his team at the University of Houston pushed this limit to almost 100 K. Finally, the liquid nitrogen temperature (77 K) was breached.
The dawn of a new era was upon us. In one year, the superconducting transition temperature climbed 70 degrees in a never before investigated class of materials. If a non-metal could superconduct at a temperature higher than the best metals, could a room-temperature superconductor be far from discovery? The euphoria of revolutionary discovery stimulated the imagination of all technologists.
The buzz 15 years ago was that high-temperature superconductors (HTS) would transform power transmission, power storage, information processing, on-chip interconnects, chip-to-chip interconnects, transportation, nuclear fusion and medical sensors. Since then, the hype has dissipated, leaving the reality behind.
The most complete Web site resource tracking superconductivity, www.superconductors.org, lists more than 30 companies actively manufacturing products leveraging both conventional low-temperature superconductors (LTS) and HTS.
Only one microelectronics application has become a key opportunity for HTS materials: high-Q tuned filters for wireless communications. Three companies manufacture and sell HTS-based rf filters: Conductus, ISCO International and Superconductor Technologies.
In second-generation (2G) and later cell phone networks, a narrow bandwidth is allocated for communications. To minimize the interference from adjacent transmission bands, filters are used in the front end of base stations. The sharper the cutoff, the more filter elements or poles are required. However, in conventional filters, with typically eight poles, the resistive loss from each pole increases the attenuation or insertion loss in the filter's bandwidth. This directly translates to a reduced range for the tower.
The use of HTS materials virtually eliminates the resistive loss in filters. HTS filters use as many as 16 poles, with no drop in insertion loss and better exclusion of interference. This property translates to a 50-80% increased range for an antenna.
Thousands of base stations with HTS filters manufactured with thick-film and thin-film technology have been shipped in the past five years. Conductus and Superconductor Technologies exclusively use thin-film technology for smaller physical size, while ISCO uses both thick- and thin-film technologies. All approaches use YBCO (yttrium barium copper oxide) materials.
Though the upper temperature limit has not budged much in the last 15 years, whole new classes of materials have been shown to be superconducting. Two years ago, C60 Buckyballs demonstrated superconductivity up to 52 K. This year, three new materials have shown superconductivity. Carbon nanotubes superconduct at the frigid temperature of only 0.5 K. However, there is the expectation that filling them with Buckyballs may dramatically increase their transition temperature. Lucent Technologies' Bell Laboratories recently announced the discovery of a plastic, polythiophene, that superconducts at nearly 40 K. Magnesium diboride, a common, innocuous metal, was found to superconduct at 39 K. This is more than twice the highest previous transition temperature of any other metal.
For additional information on assembly and packaging, go to www.semiconductor.net/assembly
