SEARCH · Engineering Papers
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Noise spectra of a InSb hot electron bolometer at 4.2 K bath temperature are measured and compared with theory. Bolometer phonon noise and Johnson noise are included in the analysis. A nonequilibrium Johnson noise model, due to Mather, gives good agreement with experiment over the frequency range 0.2-1.1 MHz. This model incorporates an elevated electron gas temperature and electrothermal feedback from the bias supply.
The principles governing the design of metal films are reviewed, with attention also given to the choice of metals. A description is then given of the characteristics of a bolometer with a gold absorbing film. It is demonstrated that gold is effective as an absorbing film for a millimeter bolometer operated at 1.5 K. At 1.5 K, gold is significantly better than bismuth since gold has a lower heat capacity for the absorbing film. At 0.3 K, gold and bismuth are both suitable. It is pointed out that at temperatures below 0.3 K, a superconducting absorbing film can have a heat capacity low enough not to dominate the heat capacity of the detector; for this reason, it may give better performance than a nonsuperconducting absorbing film.
A new type of bolometer detector for the millimeter and submillimeter spectral range is described. The bolometer is constructed of silicon using integrated circuit fabrication techniques. Ion implantation is used to give controlled resistance vs temperature properties as well as extremely low 1/f noise contacts. The devices have been tested between 4.2 and 0.3 K. The best electrical NEP measured is 4 x 10 to the -16th W/Hz to the 1/2 at 0.35 K between 1- and 10-Hz modulation frequency. This device had a detecting area of 0.25 sq cm and a time constant of 20 msec at a bath temperature of 0.35 K.
An analytical method for determining the absorbed energy calibration of an IR bolometer from purely electrical measurements is presented. The technique is targeted at detectors with purely electrical nonlinearities. It is shown that the responsivity can be derived from the I-V plot if the limiting high-frequency impedance is measured. The electrical nonlinearity is then the difference between the impedance and the dc resistance. The model is used for designing a circuit for measuring the relative impedance parameters. The results are applicable to studies of IR sources and the detection of energy pulses with bolometers.
The infrared array referred to as Big Mac (for Marshall Array Camera), was designed for ground based astronomical observations in the wavelength range 5 to 35 microns. It contains 20 discrete gallium-doped germanium bolometer detectors at a temperature of 1.4K. Each bolometer is irradiated by a square field mirror constituting a single pixel of the array. The mirrors are arranged contiguously in four columns and five rows, thus defining the array configuration. Big Mac utilized cold reimaging optics and an up looking dewar. The total Big Mac system also contains a telescope interface tube for mounting the dewar and a computer for data acquisition and processing. Initial astronomical observations at a major infrared observatory indicate that Big Mac performance is excellent, having achieved the design specifications and making this instrument an outstanding tool for astrophysics.
The first IR Detector Technology Workshop took place at NASA Ames Research Center on July 12 and 13, 1983. The conclusions presented at that meeting are still valid. More was learned about the physics of hopping conduction at very low temperatures which will be important for bolometer design and operation at ever decreasing temperatures. Resistivity measurements were extended down to 50 mK. At such low temperatures, precise knowledge of the neutron capture cross sections sigma (sub n) of the various Ge isotopes is critical if one is to make an accurate prediction of the dopant concentrations and compensation, and therefore resistivity, that will result from a given irradiation. An empirical approach for obtaining the desired resistivity material is described and the process of conducting a set of experiments which will improve the knowledge of the effective sigma (sub n) values for a given location in a particular reactor is discussed. A wider range of NTD Ge samples is now available. Noise measurements on bolometers with ion implanted contacts show the no 1/f noise component appears down to 1 Hz and probably lower.
A high-Tc superconducting bolometer has been constructed using a YBa2Cu3O(x) thin-film meander line 20 microns wide and 76,000 microns long, deposited on a SrTiO3 substrate. Radiation is absorbed by a thin film of Bi with well-characterized absorption properties deposited on a Si substrate in contact with the SrTiO3. At 1.8 Hz the measured bolometer response to a 500-K blackbody is 5.2 V/W (820 V/W extrapolated to dc). The impact of apparent nonohmic behavior at the transition is discussed, as are ways of reducing the observed 1/f noise. The response time is 32 s and is dominated by the heat capacity of the SrTiO3 substrate.
A simple polishing technique was developed for thinning the LaAlO3 substrates for high-quality Tc bolometer films, and thus reducing their heat capacity. A 150-ms bolometer was made on a LaAlO3 substrate with a 5-Hz D* value of 1.5 x 10 exp 8. It is shown that passive temperature stabilization is adequate for operation at the transition. There remained excess noise at the transition, but this noise appears to be of nonbolometric origin.
Bolometers operating at low temperatures (T greater than 0.3 K) are the best candidate for far infrared (lambda greater than or = 120 microns) and submillimeter detection from cold orbiting telescopes. The limits of sensitivity for such devices have been studied (Mather, 1983). Based on this work, we conclude that excellent sensitivity is possible given careful detector design. We have designed and are fabricating bolometers for operation at low temperatures. The crucial components in detector design are (1) heat capacity, (2) temperature sensing, and (3) absorption of infrared radiation. In this presentation, we discuss the choice of materials and fabrication techniques we have made in the interest of maximum sensitivity. In order that optimization of detector performance be reasonably easy to carry out, we have attempted to use conventional fabrication techniques wherever possible (photolithography, wire bonding). The use of such procedures allows us to make design changes on short time scales, which is essential in achieving optimal performance.
The development of arrays of infrared bolometers that are suitable for use in the Far Infrared and Submillimeter Telescope (FIRST) mission is reported. The array architecture is based on the silicon nitride micromesh bolometer currently baselined for use in the case of the Planck mission. This architecture allows each pixel to be efficiently coupled to one or both polarizations and to one or more spatial models of radiation. Micromesh structures are currently being developed, coupled with transistor-edge sensors and read out by a SQUID amplifier. If these devices are successful, then the relatively large cooling power available at 300 mK may enable a SQUID-based multiplexer to be integrated on the same wafer as the array, creating a monolithic, fully multiplexed, 2D array with relatively few connections to the sub-Kelvin stage.
Heterodyne measurements have been made at 533 GHz using a novel superconducting hot-electron bolometer in a waveguide mixer. The bolometer is a 0.3 micrometer long niobium microbridge with a superconducting transition temperature of 5 K. The short length ensures that electron diffusion dominates over electron-phonon interactions as the electron cooling mechanism, which should allow heterodyne detection with intermediate frequencies (if) of several GHz. A Y-factor response of 1.15 dB has been obtained at an if of 1.4 GHz with 77 and 295 K loads, indicating a receiver noise temperature of 650 K DSB. The -3 dB rolloff in the if response occurs at 1.7 GHz.
Hot electron bolometer (HEB) mixer elements, based on niobium titanium nitride (NbTiN) thin film technology, have been fabricated on crystalline quartz substrates over a 20 nm thick AlN buffer layer. The film was patterned by optical lithography, yielding bolometer elements that measure about 1 micrometer long and between 2 and 12 micrometers wide. These mixer chips were mounted in a fixed-tuned waveguide mixer block, and tested in the 600 and 800 GHz frequency range. The 3-dB output bandwidth of these mixers was determined to be about 2.5 GHz and we measured a receiver noise temperature of 270 K at 630 GHz using an intermediate frequency of 1.5 GHz. The receiver has excellent amplitude stability and the noise temperature measurements are highly repeatable. An 800 GHz receiver incorporating one of these mixer chips has recently been installed at the Sub-Millimeter Telescope in Arizona for field test and for astronomical observations.
Large format, two dimensional arrays of close-packed bolometers will enable submillimeter cameras and spectrometers to obtain images and spectra orders of magnitude faster than present instruments. The South Pole Imaging Fabry-Perot Interferometer (SPIFI) for the AST/RO observatory and the Submillimeter and Far-InfraRed Experiment (SAFIRE) on the SOFIA airborne observatory will employ a large-format, two-dimensional, closepacked bolometer arrays. Both these instruments are imaging Fabry-Perot spectrometers operating at wavelengths between 100 micron and 700 micron. The array format is 16x32 pixels, using a 32-element multiplexer developed in part for this purpose. The low backgrounds achieved in spectroscopy require very sensitive detectors with NEPs of order 5x10(exp 18) W/square root of Hz. Superconducting detectors can be close-packed using the Pop-Up Detector (PUD) format, and SQUID multiplexers operating at the detector bas temperature can be intimately coupled to them. We have fabricated an engineering model array with this technology which features a very compact, modular approach for large format arrays.
SHARC II is a 350 micron facility camera for the Caltech Submillimeter Observatory (CSO) expected to come on-line in 2002. The key component of SHARC II is a 12x32 array of doped silicon 'pop-up' bolometers developed at NASA/Goddard and delivered to Caltech in March 2002. Each pixel is 1 mm x 1 mm, coated with a 400 Omega/square bismuth film, and located lambda/4 above a reflective backshort to maximize radiation absorption. The pixels cover the focal plane with greater than 95% filling factor. Each doped thermistor occupies nearly the full area of the pixel to minimize 1/f noise. We report some results from the first cold measurements of this array. The bolometers were located inside a dark cover, and 4x32 pixels were read simultaneously. In the best 25% of winter nights on Mauna Kea, SHARC II is expected to have an NEFD at 350 microns of 1 Jy s(sup 1/2) or better.
A general model is presented that assimilates the thermal and electrical properties of the bolometer - this block model demonstrates the Electro-Thermal Feedback (ETF) effect on the bolometers performance. This methodology is used to construct a SPICE model that by way of analogy combines the thermal and electrical phenomena into one simulation session. The resulting circuit diagram is presented and discussed.
We discuss a new type of direct detector, a silicon hot-electron bolometer, for measurements in the far-infrared and submillimeter spectral ranges. High performance bolometers can be made using the electron-phonon conductance in heavily doped silicon to provide thermal isolation from the cryogenic bath. Noise performance is expected to be near thermodynamic limits, allowing background limited performance for many far infrared and submillimeter photometric and spectroscopic applications.
In order to provide high sensitivity rapid imaging at 3.3mm (90GHz) for the Green Bank Telescope - the world's largest steerable aperture - a camera is being built by the University of Pennsylvania, NASA/GSFC, and NRAO. The heart of this camera is an 8x8 close-packed, Nyquist-sampled detector array. We have designed and are fabricating a functional superconducting bolometer array system using a monolithic planar architecture. Read out by SQUID multiplexers, the superconducting transition edge sensors will provide fast, linear, sensitive response for high performance imaging. This will provide the first ever superconducting bolometer array on a facility instrument.
Filled arrays of planar bolometers are finding astronomical applications at wavelengths as long as several millimeters. In an effort to keep focal planes to a reasonable size while maintaining large numbers of detectors, a common strategy is to push these arrays to operate close to or at the single mode limit. Doing so introduces several new challenges that are not experienced in the multi-mode case of far-infrared detectors having similar pixel sizes. First, diffractive effects of the pixels themselves are no longer insignificant and will ultimately contribute to the resolution limit of the optical system in which they reside. We use the method of Withlngton et al. (2003) to model the polarized diffraction in this limit. Second, it is necessary to re-examine the coupling between the radiation and the absorbing element that is thermally connected to the bolometers. The small f-numbers that are often employed to make use of large focal planes makes backshort construction problematic. We introduce a new strategy to increase detector efficiency that uses an antireflective layer on the front side of the detector array. In addition, typical methods for stray light control that rely on multiple reflections in a lossy medium fail due to physical size constraints. For this application, we find that resonant absorbers are a more effective strategy that can be implemented in the space available.