Terahertz Waves for Materials Evaluation, 14-9405Printer Friendly Version
Inclusive Dates: 07/01/03 - 06/30/04
Background - The goal of this project is to develop a terahertz imaging and spectroscopy facility at Southwest Research Institute. This will be a novel non-destructive, non-ionizing imaging capability that will serve the needs of Applied Physics Division clients.
Terahertz, or submillimeter waves are electromagnetic radiation with wavelengths below about 300µm, inhabiting a "window" in the electromagnetic spectrum between infrared light and microwaves They have important potential uses in surveillance and nondestructive inspection because of their unique combination of low photon energy (non-ionizing) and penetration through most insulating materials. While there is nothing fundamentally new about submillimeter electromagnetic waves (the basic physics of their propagation has been well understood for over a century), they have been little studied until recently, in large measure because of technical difficulties in their generation and detection. Modern terahertz sources, developed during the last decade, use ultrashort (femtosecond) laser pulses to trigger wide-band photoconductive transmitters. Similar antennas are used for coherent detection. In addition, pulsed terahertz sources typically have wide bandwidth, offering the possibility to perform "time-domain" spectroscopy of materials in a spectral band that dovetails the far infrared.
Approach - We have adopted the "canonical" approach to terahertz generation and detection, pioneered by groups at AT&T Bell Labs and IBM Research. Short pulses from a titanium:sapphire femtosecond laser are incident on photoconductive switch antennas (used for generation and collection of terahertz radiation). We have incorporated novel features in our system design, including an adaptable terahertz beam delivery system, to accommodate transmission and reflection experiments; laser focusing with dispersion-compensated mirrors (instead of lenses, to prevent pulse broadening and terahertz bandwidth degradation); electro-optic detection for wide-bandwidth capability; and astigmatic focusing of the laser, to allow high terahertz power.
Accomplishments - We have designed a number of novel, micron-size photoconductive switch antennas, including a number of high-bandwidth fractal systems, and have sent these to a commercial lithographic foundry for fabrication. We have designed mechanical packages for these switches, based on commercial through-pin DIP sockets, to allow rapid interchange and testing of our antennas. The laser and terahertz beam delivery systems have been assembled. We are waiting for delivery of custom components (including nonlinear crystals for detection, and zone-refined silicon lenses for terahertz collimation) to finish setting up the facility. In the interim, we will use antenna assemblies on loan from Prof. D. Mittleman's group at Rice University to perform initial system tests. Finally, we have reassembled and tested the home-made femtosecond laser used to trigger terahertz pulses, and have added diagnostics (a home-made interferometric autocorrelator) to check its operation. It is producing 20fs pulses (the shortest in South Texas), more than adequate for use as a terahertz trigger.