Nonlinear Optical Devices: a Rainbow of Colors and Beyond

Jonathan Kurz
Department of Applied Physics
Stanford University
June 2002

Nonlinear optical devices are optical transistors, which, in electronics, are a fundamental building block because they allow one circuit to control another. Transistors make it possible to build complex and useful electrical circuits, which are then used in everything from radios to cellular phones to computers. Optical transistors hold promise for new and improved devices based on laser light. Nonlinear optical devices may soon see use in DVD players, digital movie projectors, and communication systems.

Nonlinear optical devices are special because they can change laser light from one color (or wavelength) to another. Why might we want to do this? Sometimes - for example when writing CDs and DVDs - we need a particular color. One of the reasons why a DVD stores a two-hour movie, while a CD can only hold a dozen tracks, is that a DVD laser is more orange-colored than a CD laser, which has a deep red color. Shorter wavelengths (the violet side of the rainbow) can make smaller spots on a DVD, which means that more data can be written on a single disk. Next-generation DVDs will be written with blue lasers, tripling their capacity. By the same principle, blue lasers will also lead to higher-resolution laser printers.

If no laser is readily available in the color we want, nonlinear optical devices can come to the rescue. Tomorrow's DVD players may use these devices to convert the light from a red laser to the desired blue color. Similar devices may be used to create the full spectrum of colors needed for digital movie projectors based on lasers. The ability to change the color of light is also important for optical communication systems in which each channel is assigned to a different color (just as every radio station has a unique broadcasting frequency).

Only certain materials have the nonlinear optical properties needed to make color-shifting devices. Note that glass prisms and colored cellophane do not convert light from one color to another; they merely separate white light into its constituent colors. The particular material used by my research group is a transparent crystal called lithium niobate. Because of its large-scale production and use in various optical devices, lithium niobate can be called the silicon of the optics industry. Like silicon, lithium niobate is typically available in round wafers, several inches in diameter and less than a millimeter thick, which can be processed into devices or chips.

Making a lithium niobate wafer into a nonlinear optical device is a two-step process. The wafer is first electrically shocked with a high, but carefully controlled, voltage. This step enhances its nonlinear optical properties, making color conversion more efficient. Next, waveguides are created through the wafer to guide light from one side of the device to the other in narrow channels. An optical fiber, which can confine light and guide it through twists and bends, is another an example of a waveguide. Lithium niobate waveguides are defined using many of the same tools used to make electrical circuits on silicon wafers. When laser light is sent through one of these waveguides, some or most of it is converted to another color by the time it emerges on the opposite side.

My research involves making devices with networks of waveguides. Previous devices have typically been made from a single straight waveguide, equivalent to a single optical transistor. I am developing more complicated waveguide devices that include multiple light paths: an optical circuit with the functionality of multiple optical transistors. While these devices are far from being optical computer chips (state-of-the-art computer chips contain tens of millions of transistors), they can still perform essential optical functions for future communications systems. The goal of my work is to demonstrate novel optical devices that may someday become important technology.