Rutgers Ultraviolet Dectector Lab

Testing at vacuum ultraviolet wavelengths is far more difficult than testing at visible wavelengths. Not only is there the added complexity of working blind (or partially blind) inside a vacuum chamber with limited access to optical components, most of these components are extremely sensitive to even the most minute amounts of contamination. In addition, most lamps and in some cases filters have limited lifetimes of several hundred hours. The transmission characteristics of a neutral-density filter, for example, can change by a factor of 30 in just 15 hours of operation if it is placed too close to a strong UV lamp. While none of the problems associated with vacuum ultraviolet testing are insurmountable, one requires the proper equipment and experience to insure meaningful results. Reported values are frequently found to be erroneous by factors of 2 because the appropriate care was not taken.  

Detective Quantum Efficiency (DQE) differs from ordinary QE in that a DQE measurement takes into account all system losses such as those from windows as well as an electronic conversion efficiency. In most UV applications, DQEs differ substantially from ordinary QE. For example, a CCD with a UV enhancing coating may obtain a QE of 30% but only obtains a solar-blind DQE of 2-5% at best when placed behind a woods filter to exclude the optical band.  

Rutgers University has a well equipped laboratory for vacuum UV testing. The laboratory contains 7 separate vacuum chambers ranging in size from 5 inches to 5 feet in diameter. To maximize their effectiveness, most of these chambers have been customized for specific detector evaluations, but all have standardized ports so that any of these chambers can be combined for added capability. A large variety (6 different types) of pumps along with several light sources provide low to moderate vacuums (0.001 to 0.000001 Torr) testing capability for wavelengths ranging from 5 to 700 nm. Rutgers can provide resolution testing and long-term stability testing at vacuum UV wavelengths. A key feature of our vacuum chambers are their compact size. The researcher can break vacuum and be back down to operating vacuum in a few minutes. Most facilities at other universities have large, general-purpose vacuum chambers requiring much longer pump down times. The compactness of our chambers permits us to test hundreds of devices per year.  

The laboratory has 2 standard transfer sensors that are calibrated on alternate years by the National Institute of Standards (NIST). These sensors have wavelength coverages of 115 to 254 nm and 5 to 254 nm. Three additional secondary reference standards are routinely cross correlated with each other and the NIST transfer standards.  

The laboratory has 3 monochromators, an Acton 502 for wavelength coverage from 120 to 300 nm, a McPherson 247 for 5 to 120 nm, and an Oriel for 200 to 600 nm. The later is used for solar blind tests down to the one part per million level. The McPherson 247 is used in a windowless environment and uses a hallow cathode lamp.  

A small DQE chamber, pictured elsewhere, measure vacuum DQEs over wavelengths between 1200 and 3000 Angstroms. There is some 4 orders of magnitude difference between the sensitivity of the NIST standard photodiode and the detector. To accommodate this large difference, most setups for measuring UV DQEs require the introduction of some attenuating optical component with its intrinsic calibration uncertainties. Our chamber employs an electrometer with an exceptionally low noise threshold and a PMT with a very large dynamic range. These features permit a simple transfer from the NIST standard diode to the PMT and then another simple transfer from the PMT to the detector, avoiding this extra optical element and one significant source of uncertainty found in most other measurements.  

Another chamber known as the Flat-Field Chamber serves 3 important purposes: 1) pixel-to-pixel uniformity measurements, 2) long-term (30-day) flat-field stability evaluations, and 3) to determine the pixel-to-pixel uniformity as a function of wavelength. The chamber can select from one of 6 wavelength band-pass filters ranging from 1216 to 2800 Angstroms and one of 7 neutral-density filters ranging from open to ND4. The lambertian error, a measurement of the uniformity, is less than 2% over the surface of the detector. Initial tests with its STIS predecessor chamber in 1992 resulted in a 30-fold degradation in the transmission of one ND filter in a 24 hour interval of operation. As a consequence, a spacing flange was introduced to move the D2 lamp further away from the filters. Subsequently, the Flat-Field Stability Chamber has been demonstrated to maintain repeatability to 0.5% over periods greater than a month.  

A third chamber is used for detailed, high resolution measurements. It is capable of performing Modulus Transfer Function tests at 121.6 nm on a spatial scale of 10 microns. This chamber is also being made ready for testing of wafer devices that have not yet been packaged. IC probes connected to shielded feedthroughs, translating tables and other equipment are being fitted to this chamber for the purpose of vacuum testing the various devices.  

In-air DQE measurements are made using a matched Oriel lamp housing plus monochromator, which sends a selected wavelength (color) of light into a light tight box. Many astronomical objects emit 10^4 to 10^8 visible photons for every ultraviolet photon. If the UV detector has any sensitivity to visible light, the detector will be swamped by the visible light. Inside the light-tight box the signal from the detector of interest is measured against the signal produced by a standard, calibrated detector. Most NASA UV space missions require a detector to be solar blind (insensitive) by a factor of at least a million to visible light. The Rutgers laboratory can place the beam directly onto the detector plus comparison or can use a fold mirror for table operation. In the latter case, many immature devices can be tested using micropositioners to make electronic contact on wafer fragments. The laboratory has 3 electrometers, two capable of picoamp and one capable of femtoamp operation.  

Auxiliary equipment includes a Electro-Static Discharge (EDS) station and an environmentally-controlled storage area. Many devices such as CCDs can be destroyed by static discharges that are 100 times smaller than can be felt my humans. We have the capability to store as well as handle these devices in a dry, EDS-free environment. The laboratory also has an optical chopper (for in-air use only), useful for determining electronic time constants.    

Non-Disclosure Policy

Rutgers has adopted a strict policy of non-disclosure of test results, except those deemed by all parties to be worthy of publication. Our test and evaluation facility supports engineering development efforts which inherently run into difficulties from time to time. It is our desire to provide reliable feedback to these investigators to increase their progress. Premature, negative publicity has the potential to kill good viable technologies before these can be perfected.

Back to Lab Home Page