SPIE Digital Library Get updates from SPIE Newsroom
  • Newsroom Home
  • Astronomy
  • Biomedical Optics & Medical Imaging
  • Defense & Security
  • Electronic Imaging & Signal Processing
  • Illumination & Displays
  • Lasers & Sources
  • Micro/Nano Lithography
  • Nanotechnology
  • Optical Design & Engineering
  • Optoelectronics & Communications
  • Remote Sensing
  • Sensing & Measurement
  • Solar & Alternative Energy
  • Sign up for Newsroom E-Alerts
  • Information for:
    Advertisers
SPIE Defense + Commercial Sensing 2017 | Call for Papers

Journal of Medical Imaging | Learn more

SPIE PRESS




Print PageEmail PageView PDF

Optoelectronics & Communications

Laying it on thick

Eye on Technology - integrated optics

From oemagazine December 2001
30 December 2001, SPIE Newsroom. DOI: 10.1117/2.5200112.0003

With the metropolitan-area-network (MAN) market widely hailed as the next big growth area for optical communications, cost reduction has become a paramount issue for components manufacturers. One avenue for accomplishing this task is chip-level integration of passives, or of actives with passives. From a manufacturing point of view, coating an entire silicon wafer with silica can be challenging, with techniques like flame hydrolysis deposition and chemical vapor deposition (CVD) requiring high temperatures and extended times for deposition. At the European Conference on Optical Communications (ECOC; Amsterdam, The Netherlands; 1–5 October), Heriot-Watt University (Edinburgh, UK) spinout Terahertz Photonics (TP; Livingston, Scotland, UK) offered a fast, low-temperature alternative—a sol-gel method for silica-on-silicon circuits.

Sol-gel refers to a technique in which a solid material is produced from a room-temperature suspension called a sol. The sol is deposited, dried to remove liquids from within the matrix, then sintered to fill the pores in the matrix. On the face of it, spin-coating a sol-gel layer of silica on a silicon substrate seems simple. In the past, however, internal stresses in the silica film introduced during drying and densification caused the films to crack or craze, particularly when they were thicker than 1 µm, making sol-gel unsuitable for optoelectronic applications.

TP seems to have solved the problem of cracks. Using only a basic spinner and a furnace, engineers deposit silica films up to 8-µm thick in a single step, producing a waveguide substrate compatible with low-loss, single-mode transmission. "The secret is in the chemistry," says Navin Suyal, director of the PLC division at TP. "The structure and the density of the SiO2 units floating in the sol determine whether or not the film will withstand the capillary forces during the drying. Our sol synthesis and modification process controls these properties of silica units. Therefore, the organic can be eliminated during the drying and densification even when the films are very thick, leaving a uniform, continuous, and good-quality film of silica." A subsequent high-temperature anneal densifies the gel film to an optical-quality glass film. The company is presently working with 4-in. substrates, but Suyal says they have also demonstrated the process for 6-in. substrates.

"Sol-gel has been around for many years and has a 10-year history with optics," says Professor Richard de la Rue of the University of Glasgow (Glasgow, Scotland, UK). The key question, he says, is whether TP has made the crack-free thin films that yield low-loss waveguides. "If they have, then the prospects for these components are good. Such devices are going to be needed in the wider spread of fiber networks." On-chip integration is not an easy goal to accomplish, he points out. "The losses per component are going to be key. Added up, the losses could overwhelm the system; the decibel per component could be the limiting factor. Interfaces are the killer in optics, and fiber-to-chip alignment is also important. [The TP process] could play a key role in achieving the tough requirements in this field."

The silica-on-silicon market is awash with new companies, says Andrew Fletcher of Reed Electronics Research (Sutton, UK). "There are quite a few players worldwide with their eyes on this market," he says. "For sure, not all will survive the industry adjustments that will be an inevitable part of the industry upturn next year." There is a place for companies that offer more than one materials system, Fletcher adds. "No single materials system, be it silicon-on-insulator, silica-on-silicon, or even indium phosphide (InP) will satisfy all the needs of optics devices for datacom network devices. Nor can these materials be combined easily, but this will come in due course. Within the next five years, hybrid solutions combining all three of these and maybe others will make an appearance."

Indeed, TP is also developing a polymer-on-silicon product designed for datacom and metro/access loop applications. The halogenated polymers are stable to 300°C and offer losses of less than 0.5 dB/cm at 1550 nm. With a high coefficient of thermal expansion, the polymer-coated wafers offer advantages for certain applications.