ULTRAVIOLET DETECTORS
UV SURF is up at NIST
GARY T. FORREST
The upgraded Synchrotron Ultraviolet Radiation Facility (SURF III) at the National Institute for Standards and Technology (NIST; Gaithersburg, MD) was scheduled for completion by the end of 1998. The major upgrade has targeted precision calibration of ultraviolet (UV) detectors and measurements of multilayer thin-film coatings.
To achieve absolute precision of better than 1%, major improvements in magnetic-field uniformity have been achieved--from the approximately 1% of SURF II to better than 0.04% (see photo on p. 54). According to NIST researcher Ping-Shine Shaw, achieving the improved magnetic-field uniformity and knowing the circulating current precisely will allow the energy distribution of the resulting electron beam to be calculated precisely. From this value an exact calculation of the synchrotron output can be made.
SURF III will be used to calibrate detectors by comparing the responsivity of the detector as a function of wavelength to the calculated energy distribution of SURF III. Depending on the beam energy, wavelengths from 100 µm to 10 nm are accessible. Of particular note will be the ability of this relatively low-energy storage ring to calibrate detectors in the 125- to 320-nm UV region. Classic blackbody radiation sources--the primary standard for most detector calibration--have very little energy at short wavelengths. Several beam lines are being installed for various uses including multilayer UV coating evaluation, use of an absolute cryogenic radiometer for cross comparison with existing NIST detector-calibration systems, and other calibration measurements.
Other testbeds
Three other UV testbeds are also beginning full operation this year. These include precision refractive-index measurements of fused silica and calcium fluoride at 193 and 157 nm in support of advanced lithography-stepper-lens design. A precision UV Fourier-transform interferometer has been constructed to extend the 10 parts-per-million (ppm) refractive-index measurement made previously to 1 ppm. Similarly, a MPB Technologies (Pointe Claire, Que., Canada) excimer laser producing 5 mJ/cm2 has been incorporated into a detector testbed to evaluate detector damage at 193 and 157 nm. Researcher Rajeev Gupta said the system is capable of measuring both detector responsivity and reflectivity so that surface and bulk-damage mechanisms can be evaluated, thereby allowing UV-damage-resistant detectors to be designed and tested.
Broadly tunable laser systems based on Ti:sapphire and periodically poled lithium niobate (PPLN) have also been assembled. According to researcher Keith Lykke, the SIRCUS (Spectral Irradiance and Radiance Calibration using Uniform Sources) is designed to provide narrow-bandwidth, high-intensity radiation from 200 nm to 20 µm. The goal is to offer absolute detector calibration in both radiance and irradiance. To achieve uniform beams with suitable point source and Lambertian spatial distributions, 1 mW at the test wavelengths is required at the input to custom integrating spheres.
For wavelengths from 200 to 1100 nm, the system uses optical buildup cavities to achieve high continuous-wave (CW) doubling efficiencies without requiring high-peak-power pulsed operation. This will allow detectors to be calibrated with CW beams and without regard to potential nonlinear effects that might accompany the use of a modelocked pulse train.
The results of detector illumination are then compared with a high-accuracy cryogenic radiometer, achieving microwatt sensitivity for absolute measurements. This same absolute standard is used with the second mid-infrared laser system based on PPLN, which is tunable initially from 1.4 to 2.1 µm and 2.1 to 4.7 µm. Upgrades, mixing in silver gallium selenide (AgGaSe), will allow wavelengths as long as 20 µm to be reached. To achieve sufficiently high peak powers, this system is expected to be operated in a modelocked configuration. The PPLN system has been assembled and preliminary characterization of its performance completed. By the end of 1998, the PPNL system and basic 200--1100-nm systems will be operational with continuing upgrades for the next two years.
Gary T. Forrest
GARY T. FORREST is president of SensorPhysics, Oldsmar, FL; www.sensorphysics.com.