Technology Focus Electronics/Computers · 11 Photodetector Arrays for Multicolor Visible/Infrared Imaging 12 Semiconductor Bolometers Give Background- ... 22 Solar-Powered Cooler

Technology Focus

Electronics/Computers

Software

Materials

Mechanics

Machinery/Automation

Manufacturing & Prototyping

Bio-Medical

Physical Sciences

Information Sciences

Books and Reports

10-06 October 2006

https://ntrs.nasa.gov/search.jsp?R=20110013574 2020-06-17T06:40:11+00:00Z

NASA Tech Briefs, October 2006 1

INTRODUCTIONTech Briefs are short announcements of innovations originating from research and develop-

ment activities of the National Aeronautics and Space Administration. They emphasizeinformation considered likely to be transferable across industrial, regional, or disciplinary linesand are issued to encourage commercial application.

Availability of NASA Tech Briefs and TSPsRequests for individual Tech Briefs or for Technical Support Packages (TSPs) announced herein shouldbe addressed to

National Technology Transfer CenterTelephone No. (800) 678-6882 or via World Wide Web at www2.nttc.edu/leads/

Please reference the control numbers appearing at the end of each Tech Brief. Information on NASA’s Innovative Partnerships Program (IPP), its documents, and services is also available at the same facility oron the World Wide Web at http://ipp.nasa.gov.

Innovative Partnerships Offices are located at NASA field centers to provide technology-transfer access toindustrial users. Inquiries can be made by contacting NASA field centers listed below.

Ames Research CenterLisa L. Lockyer(650) [email protected]

Dryden Flight Research CenterGregory Poteat(661) [email protected]

Goddard Space Flight CenterNona Cheeks(301) [email protected]

Jet Propulsion LaboratoryKen Wolfenbarger(818) [email protected]

Johnson Space CenterMichele Brekke(281) [email protected]

Kennedy Space CenterDavid R. Makufka(321) [email protected]

Langley Research CenterMartin Waszak(757) [email protected]

Glenn Research CenterRobert Lawrence(216) [email protected]

Marshall Space Flight CenterVernotto McMillan(256) [email protected]

Stennis Space CenterJohn Bailey(228) 688-1660 [email protected]

Carl Ray, Program ExecutiveSmall Business Innovation Research (SBIR) & Small Business Technology Transfer (STTR) Programs(202) [email protected]

Merle McKenzieInnovative Partnerships Program Office(202) [email protected]

NASA Field Centers and Program Offices

5 Technology Focus: Sensors5 Protein Sensors Based on Optical Ring Resonators

6 Phase Sensor for Aligning a Segmented TelescopeMirror

6 Control Software for Advanced Video GuidanceSensor

6 Generating Control Commands From GesturesSensed by EMG

9 Electronics/Computers9 Multiple-Flat-Panel System Displays

Multidimensional Data

10 3D X-Ray Luggage-Screening System

11 Probe Station and Near-Field Scanner for TestingAntennas

11 Photodetector Arrays for MulticolorVisible/Infrared Imaging

12 Semiconductor Bolometers Give Background-Limited Performance

13 Multichannel X-Band Dielectric-ResonatorOscillator

14 Automatic Alignment of Displacement-MeasuringInterferometer

15 Software15 Earth Observing System Data Gateway

15 Power User Interface

15 Mercury Shopping Cart Interface

15 Cassini Archive Tracking System

16 Architecture Adaptive Computing Environment

16 Computing Fault Displacements From SurfaceDeformations

17 Materials17 Oxygen-Permeable, Hydrophobic Membranes of

Silanized α-Al2O3

17 SiC Composite Turbine Vanes

19 Mechanics19 Retaining Device for the Interior Structure of a

Spacecraft Payload

19 Tool for Torquing Circular Electrical-ConnectorCollars

21 Machinery/Automation21 System for Continuous Deaeration of

Hydraulic Oil

22 Solar-Powered Cooler and Heater for anAutomobile Interior

23 Manufacturing & Prototyping23 Improved Oxygen-Beam Texturing of Glucose-

Monitoring Optics

23 Tool for Two Types of Friction Stir Welding

25 Bio-Medical25 Stationary Apparatus Would Apply Forces of

Walking to Feet

25 Instrument Would Detect and Collect BiologicalAerosols

27 Physical Sciences27 Boundary Condition for Modeling Semiconductor

Nanostructures

28 Miniature Distillation Column for Producing LOXFrom Air

29 Even Illumination From Fiber-Optic-Coupled LaserDiodes

30 Optically Driven Deformable Mirrors

31 Information Sciences31 Algorithm for Automated Detection of Edges of

Clouds

31 Exploiting Quantum Resonance to SolveCombinatorial Problems

31 Hybrid Terrain Database

33 Books & Reports33 On Release of Microbe-Laden Particles From Mars

Landers

33 A Concept for Run-Time Support of the ChapelLanguage

33 Thermoelectric Inhomogeneities in(Ag1–ySbTe2)x(PbTe)1–x

33 Spacecraft Escape Capsule

10-06 October 2006


This document was prepared under the sponsorship of the National Aeronautics and Space Administration. Neither the United States Govern-ment nor any person acting on behalf of the United States Government assumes any liability resulting from the use of the information containedin this document, or warrants that such use will be free from privately owned rights.


Protein Sensors Based on Optical Ring ResonatorsProgress has been achieved in the continuing development of optical chemical sensors.NASA’s Jet Propulsion Laboratory, Pasadena, California

Prototype transducers based on inte-grated optical ring resonators have beendemonstrated to be useful for detectingthe protein avidin in extremely dilute so-lutions. In an experiment, one of thetransducers proved to be capable of indi-cating the presence of avidin at a con-centration of as little as 300 pM in abuffer solution — a detection sensitivitycomparable to that achievable by previ-ously reported protein-detection tech-niques. These transducers are serving asmodels for the further development ofintegrated-optics sensors for detectingsmall quantities of other proteins andproteinlike substances.

The basic principle of these transduc-ers was described in “Chemical SensorsBased on Optical Ring Resonators”(NPO-40601), NASA Tech Briefs, Vol. 29,No. 10 (October 2005), page 32. Thedifferences between the present trans-ducers and the ones described in thecited prior article lie in details of imple-mentation of the basic principle. As be-fore, the resonator in a transducer ofthe present type is a closed-circuit di-electric optical waveguide. The outer-most layer of this waveguide, analogousto the optical cladding layer on an opti-cal fiber, consists of a layer comprisingsublayers having indices of refractionlower than that of the waveguide core.The outermost sublayer absorbs thechemical of interest (in this case,avidin). The index of refraction of theoutermost sublayer changes with theconcentration of absorbed avidin. Theresonator is designed to operate withrelatively strong evanescent-wave cou-pling between the outer sublayer andthe electromagnetic field propagatingalong the waveguide core. By virtue ofthis coupling, the chemically inducedchange in the index of refraction of theoutermost sublayer causes a measurablechange in the spectrum of the resonatoroutput.

The figure depicts one of the proto-type transducers, wherein the ring res-onator is a dielectric optical waveguidelaid out along a closed path resemblinga racetrack. The waveguide includes a

core of SixNy formed on an innercladding layer of SiO2 on a substrate ofSi. The outer cladding layer comprisesan inner sublayer of SiO2 and an outersublayer of biotin. (The SiO2 sublayer isneeded for binding the biotin to theSixNy core.) The selectivity of the sensordepends on the use of biotin, whichbinds specifically to avidin, immobilizingavidin on the outer surface and therebychanging the index of refraction. Theportion of the cross section occupied bythe propagating electromagnetic modeis confined laterally by the rib portion ofthe core and is shown in the figure as anoval. In addition to the ring resonator,there are straight input and output wave-guides separated from the straight seg-ments of the ring resonator by anevanescent-wave-coupling gap of 1.6 µm.

In operation, the transducer ismounted in a flow cell on a copperchuck. The temperature of the chuck(and, thus, of the transducer) is moni-tored by use of a thermistor and con-trolled by use of a thermoelectric cooler.A solution containing avidin is pumpedthrough the flow cell. Through thestraight input waveguide, the resonatoris illuminated at a wavelength of 633 nmby a He-Ne laser. The length of theclosed optical path of the resonator ringvaries with the temperature, and the

temperature is adjusted to keep the pathlength an integer multiple of a wave-length: that is, the temperature is ad-justed to maintain operation at one ofthe resonances. As the biotin coating ab-sorbs avidin, the resulting change in theindex of refraction manifests itself as achange in the resonance wavelengthand, hence, in the temperature neededto maintain the chosen resonance.Hence, further, the change in the con-trolled temperature can be taken as anindication of the amount of dissolvedavidin to which the transducer has beenexposed.

This work was done by Ying Lin andAlexander Ksendzov of Caltech for NASA’sJet Propulsion Laboratory. Further infor-mation is contained in a TSP (see page 1).

In accordance with Public Law 96-517,the contractor has elected to retain title to thisinvention. Inquiries concerning rights for itscommercial use should be addressed to:

Innovative Technology Assets ManagementJPLMail Stop 202-2334800 Oak Grove DrivePasadena, CA 91109-8099(818) 354-2240E-mail: [email protected]

Refer to NPO-41585, volume and number ofthis NASA Tech Briefs issue, and the pagenumber.

25 Å

130 Å1.5 µm 0.4 mm

Radius = 2 mm

Input

ENLARGED CROSS SECTION OFRING RESONATOR WAVEGUIDE

PLAN VIEW OF RINGRESONATOR AND INPUT AND

OUTPUT WAVEGUIDES

Output B

Output A

Notes:1. Not to scale2. n denotes index of refraction.

1.8 µm

SixNy Core,n = 1.884

Biotin Outer Sublayer of Outer Cladding Layer

SiO2 Inner Sublayer of Outer Cladding Layer

Si Substrate Si Substrate

SiO2 Inner Cladding,n = 1.467

NPO- 41585ABPI

9-8-05 bs

Technology Focus: Sensors

A Biotin-Clad Optical Ring Resonator acts as an avidin sensor in that the resonance spectrum becomesshifted in wavelength when the biotin absorbs avidin.

6 NASA Tech Briefs, October 2006

A phase sensor has been developed foruse in aligning a segmented telescopemirror to within a fraction of a wave-length in piston. (As used here, “piston”signifies displacement of a mirror seg-ment along the optical axis of the tele-scope.) Such precise alignment is neces-sary in order to realize the full benefit ofthe large aperture achievable throughsegmentation.

This phase sensor is achromatic. It isbased on two-wavelength shearing interfer-ometry, and can be modified to utilize anextended or broad-band (e.g., white) lightsource. The sensor optics include a ruleddiffraction grating and an imaging lens.

The sensor can measure the pistonshift between segments as well as aberra-tions of the segments. It can measure

the surface error of an individual seg-ment, making it possible to compensatefor the error with optimal amount(s) ofpiston and/or tilt. The precise capturerange of the sensor depends partly onthe telescope design; the largest relativepiston shifts measurable by use of thissensor are of the order of 100 µm. Theaccuracy of the sensor also dependspartly on the telescope design; in gen-eral, the accuracy is sufficient to enablealignment to within approximately halfa wavelength. The interferometricimage is digitized and processed by asimple algorithm in real time, and theoutput of the algorithm can be used tomaintain alignment in real time, even inthe presence of atmospheric turbu-lence.

The sensor is robust. Through calibra-tion, it can be made insensitive to (and,hence, tolerant of) misalignments andaberrations of its own optics, most aber-rations of the telescope as a whole (incontradistinction to aberrations of indi-vidual segments), and most aberrationsintroduced by atmospheric turbulence.

This work was done by H. Philip Stahlof Marshall Space Flight Center andChanda Bartlett Walker of Pace & Waite,Inc. Further information is contained in aTSP (see page 1).

This invention is owned by NASA, and apatent application has been filed. For furtherinformation, contact Sammy Nabors, MSFCCommercialization Assistance Lead, [email protected]. Refer to MFS-31852-1.

Phase Sensor for Aligning a Segmented Telescope MirrorAlignment can be maintained even in the presence of atmospheric turbulence.Marshall Space Flight Center, Alabama

Embedded software has been devel-oped specifically for controlling an Ad-vanced Video Guidance Sensor (AVGS).[As described in several previous NASATech Briefs articles, a Video GuidanceSensor is an optoelectronic system thatprovides guidance for automated dock-ing of two vehicles (spacecraft in theoriginal intended application). Such asystem includes pulsed laser diodes anda video camera, the output of which isdigitized. From the positions of digitizedtarget images and known geometric rela-tionships, the relative position and ori-entation of the vehicles are computed.]The present software consists of two sub-

programs running in two processors thatare parts of the AVGS. The subprogramin the first processor receives commandsfrom an external source, checks thecommands for correctness, performscommanded non-image-data-processingcontrol functions, and sends image-data-processing parts of commands to thesecond processor. The subprogram inthe second processor processes imagedata as commanded. Upon power-up,the software performs basic tests of func-tionality, then effects a transition to astandby mode. When a command is re-ceived, the software goes into one of sev-eral operational modes (e.g. acquisition

or tracking). The software then returns,to the external source, the data appro-priate to the command.

This program was written by Richard T.Howard, Michael L. Book, and Thomas C.Bryan of Marshall Space Flight Center.Further information is contained in a TSP(see page 1).

This invention has been patented by NASA(U.S. Patent No. 6,888,476). Inquiries con-cerning nonexclusive or exclusive license for itscommercial development should be addressed toSammy Nabors, MSFC Commercialization As-sistance Lead, at [email protected] to MFS-31865-1.

Control Software for Advanced Video Guidance SensorMarshall Space Flight Center, Alabama

Generating Control Commands From Gestures Sensed by EMGElectrical signals from muscles involved in gestures are recognized.Ames Research Center, Moffett Field, California

An effort is under way to develop nonin-vasive neuro-electric interfaces throughwhich human operators could control sys-tems as diverse as simple mechanical de-vices, computers, aircraft, and even space-

craft. The basic idea is to use electrodes onthe surface of the skin to acquire elec-tromyographic (EMG) signals associatedwith gestures, digitize and process theEMG signals to recognize the gestures,

and generate digital commands to per-form the actions signified by the gestures.

In an experimental prototype of suchan interface, the EMG signals associatedwith hand gestures are acquired by use


of several pairs of electrodes mounted insleeves on a subject’s forearm (see fig-ure). The EMG signals are sampled anddigitized. The resulting time-series dataare fed as input to pattern-recognitionsoftware that has been trained to distin-guish gestures from a given gesture set.The software implements, among otherthings, hidden Markov models, whichare used to recognize the gestures asthey are being performed in real time.

Thus far, two experiments have beenperformed on the prototype interface todemonstrate feasibility: an experiment insynthesizing the output of a joystick andan experiment in synthesizing the outputof a computer or typewriter keyboard. Inthe joystick experiment, the EMG signalswere processed into joystick commandsfor a realistic flight simulator for an air-plane. The acting pilot reached out intothe air, grabbed an imaginary joystick,and pretended to manipulate the joy-stick to achieve left and right banks andup and down pitches of the simulatedairplane. In the keyboard experiment,the subject pretended to type on a nu-merical keypad, and the EMG signalswere processed into keystrokes.

The results of the experimentsdemonstrate the basic feasibility of thismethod while indicating the need forfurther research to reduce the incidenceof errors (including confusion amonggestures). Topics that must be addressedinclude the numbers and arrangementsof electrodes needed to acquire suffi-cient data; refinements in the acquisi-

tion, filtering, and digitization of EMGsignals; and methods of training the pat-tern-recognition software.

The joystick and keyboard simula-tions were chosen for the initial experi-ments because they are familiar to manycomputer users. It is anticipated that, ul-timately, interfaces would utilize EMGsignals associated with movements morenearly natural than those associatedwith joysticks or keyboards. Future ver-sions of the pattern-recognition soft-ware are planned to be capable ofadapting to the preferences and day-to-day variations in EMG outputs of indi-vidual users; this capability for adapta-tion would also make it possible to

select gestures that, to a given user, feelthe most nearly natural for generatingcontrol signals for a given task (pro-vided that there are enough properlypositioned electrodes to acquire theEMG signals from the muscles involvedin the gestures).

This work was done by Kevin R. Wheelerand Charles Jorgensen of Ames ResearchCenter. Further information is containedin a TSP (see page 1).

This invention has been patented by NASA(U.S. Patent No. 6,720,984). Inquiries con-cerning rights for the commercial use of thisinvention should be addressed to the AmesTechnology Partnerships Division at (650)604-2954. Refer to ARC-14494-1.

Electrodes on a Subject's Forearm were used to acquire EMG signals in experiments on synthesizingjoystick attitude controls for simulated airplane flight and synthesizing typing on a numerical keypad.

ARC-14494-1 ABPI

6-16-06 em


Electronics/Computers

Multiple-Flat-Panel System Displays Multidimensional Data Related images are displayed simultaneously to facilitate perception of trends in data.Ames Research Center, Moffett Field, California

The NASA Ames hyperwall is a displaysystem designed to facilitate the visuali-zation of sets of multivariate and multi-dimensional data like those generatedin complex engineering and scientificcomputations. The hyperwall includes a77 matrix of computer-driven flat-panelvideo display units, each presenting animage of 1,280×1,024 pixels. The term“hyperwall” reflects the fact that this sys-tem is a more capable successor to priorcomputer-driven multiple-flat-panel dis-play systems known by names that in-clude the generic term “powerwall” andthe trade names “PowerWall” and “Pow-erwall.”

Each of the 49 flat-panel displays isdriven by a rack-mounted, dual-central-processing-unit, workstation-class per-sonal computer equipped with a high-performance graphical-display circuitcard and with a hard-disk drive having astorage capacity of 100 GB. Each suchcomputer is a slave node in a master/slave computing/data-communicationsystem (see Figure 1). The computerthat acts as the master node is similar tothe slave-node computers, except that itruns the master portion of the systemsoftware and is equipped with a key-board and mouse for control by ahuman operator. The system utilizescommercially available master/slavesoftware along with custom softwarethat enables the human controller to in-teract simultaneously with any numberof selected slave nodes.

In a powerwall, a single rendering taskis spread across multiple processors andthen the multiple outputs are tiled intoone seamless superdisplay. It must benoted that the hyperwall concept sub-sumes the powerwall concept in that asingle scene could be rendered as a mo-saic image on the hyperwall. However,the hyperwall offers a wider set of capa-bilities to serve a different purpose: Thehyperwall concept is one of (1) simulta-neously displaying multiple different butrelated images, and (2) providing meansfor composing and controlling such setsof images. In place of elaborate softwareor hardware crossbar switches, the hy-

ARC-15037 Fig 1ABPI

11-18-03 es

MasterComputer

Slave Computer(One of 49)

Graphics Card(One of 50)

Liquid-CrystalDisplay Panel(One of 50)

Keyboard Mouse

EthernetSwitch × 7 × 7

Figure 1. Master and Slave Nodes Communicate via a high-speed local-area network.

ARC-15037 Fig 2 ABPI

050-31-06 le

Figure 2. Each of 16 Images displayed on a hyperwall represents results of a computational simulationof airflow about an aerospacecraft (specifically, a proposed reusable launch vehicle) at one of 16 dif-ferent combinations of Mach numbers and angles of attack.


perwall concept substitutes reliance onthe human visual system for integration,synthesis, and discrimination of patternsin complex and high-dimensional dataspaces represented by the multiple dis-played images.

The variety of multidimensional datasets that can be displayed on the hyper-wall is practically unlimited. For exam-ple, Figure 2 shows a hyperwall display ofsurface pressures and streamlines from acomputational simulation of airflow

about an aerospacecraft at various Machnumbers and angles of attack. In this dis-play, Mach numbers increase from left toright and angles of attack increase frombottom to top. That is, all images in thesame column represent simulations atthe same Mach number, while all imagesin the same row represent simulations atthe same angle of attack. The same view-ing transformations and the same map-ping from surface pressure to colorswere used in generating all the images.

This work was done by Daniel Gundo andCreon Levit of Ames Research Center;Christopher Henze, Timothy Sandstrom,David Ellsworth, and Bryan Green of Ad-vanced Management Technology, Inc.; andArthur Joly of Computer Science Corporation.Further information is contained in a TSP(see page 1).

Inquiries concerning rights for the commer-cial use of this invention should be addressedto the Ames Technology Partnerships Divisionat (650) 604-2954. Refer to ARC-15037-1.

3D X-Ray Luggage-Screening System3D displays would help inspectors distinguish among objects at different depths.Marshall Space Flight Center, Alabama

A three-dimensional (3D) x-ray lug-gage-screening system has been pro-posed to reduce the fatigue experiencedby human inspectors and increase theirability to detect weapons and other con-traband. The system and variantsthereof could supplant thousands of x-ray scanners now in use at hundreds ofairports in the United States and othercountries. The device would be applica-ble to any security checkpoint applica-tion where current two-dimensionalscanners are in use.

A conventional x-ray luggage scannergenerates a single two-dimensional (2D)image that conveys no depth informa-tion. Therefore, a human inspectormust scrutinize the image in an effort tounderstand ambiguous-appearing ob-jects as they pass by at high speed on aconveyor belt. Such a high level of con-centration can induce fatigue, causingthe inspector to reduce concentrationand vigilance. In addition, because ofthe lack of depth information, contra-band objects could be made more diffi-cult to detect by positioning them nearother objects so as to create x-ray imagesthat confuse inspectors.

The proposed system would make itunnecessary for a human inspector to in-terpret 2D images, which show objects atdifferent depths as superimposed. In-stead, the system would take advantageof the natural human ability to infer 3Dinformation from stereographic orstereoscopic images. The inspectorwould be able to perceive two objects atdifferent depths, in a more nearly natu-ral manner, as distinct 3D objects lying atdifferent depths. Hence, the inspectorcould recognize objects with greater ac-curacy and less effort.

The major components of the pro-posed system would be similar to those ofx-ray luggage scanners now in use. As in aconventional x-ray scanner, there wouldbe an x-ray source. Unlike in a conven-tional scanner, there would be two x-rayimage sensors, denoted the left and rightsensors, located at positions along theconveyor that are upstream and down-stream, respectively (see figure). X-ray il-lumination may be provided by a singlesource or by two sources. The position ofthe conveyor would be detected to pro-vide a means of matching the appropri-ate left- and right-eye images of an itemunder inspection.

The appropriate right- and left-eye im-ages of an item would be displayed simulta-neously to the right and left eyes, respec-tively, of the human inspector, using

commercially available stereo displayscreens. The human operator could adjustviewing parameters for maximum viewingcomfort. The stereographic images thusgenerated would differ from true stereo-scopic images by small distortions that arecharacteristic of radiographic images ingeneral, but these distortions would not di-minish the value of the images for identify-ing distinct objects at different depths.

This work was done by Kenneth Fernandez ofMarshall Space Flight Center. Further infor-mation is contained in a TSP (see page 1).

This invention has been patented by NASA(U.S. Patent No. 6,763,083 B2). Inquiries con-cerning nonexclusive or exclusive license for itscommercial development should be addressed toSammy Nabors, MSFC Commercialization As-sistance Lead, at [email protected] to MFS-31783.

MFS-31783ABPI

08-18-05 LE

Track

Conveyor Belt

Luggage

Right Sensor Left Sensor

Drive Roller

Drive Roller

Common AlignmentLine

Left X-Ray-Emitting

Tube

Right X-Ray-Emitting

Tube

Images From the Left- and Right-Eye X-Ray Sensors are captured at different locations and storedalong with conveyor position data allowing the appropriate left- and right-eye images to be viewedby the human operator on a commercially available stereo display screen.


A facility that includes a probe stationand a scanning open-ended waveguideprobe for measuring near electromag-netic fields (see figure) has been addedto Glen Research Center’s suite of an-tenna-testing facilities, at a small frac-tion of the cost of the other facilities.This facility is designed specifically fornondestructive characterization of theradiation patterns of miniaturized mi-crowave antennas fabricated on semi-conductor and dielectric wafer sub-strates, including active antennas thatare difficult to test in traditional an-tenna-testing ranges because of fragility,smallness, or severity of DC-bias or test-fixture requirements. By virtue of thesimple fact that a greater fraction of ra-diated power can be captured in a near-field measurement than in a conven-tional far-field measurement, thisnear-field facility is convenient for test-ing miniaturized antennas with lowgains.

This facility makes it possible to test acomplete set or any subset of a multi-plicity of antennas on the same sub-strate in one session. The multiple an-tennas can all be of the same design ordifferent designs. Unlike in prior an-tenna-testing facilities, there is no needfor wafer-level dicing or packaging toisolate individual antennas from a mul-tiple-antenna substrate before testing,and no need for special text fixtures.Hence, alternative prototype antennadesigns can be evaluated in rapid suc-cession to converge on an optimum de-sign in less time (and, hence, at less

cost) than in prior antenna-testing facil-ities.

In this facility, radio-frequency (RF) sig-nals and DC bias voltages and currents aresupplied to an antenna under test (AUT)through RF and DC probes, respectively,that are parts of the probe station. Theequipment in this facility includes a com-mercially available RF probe station, acoplanar-waveguide ground-signal-groundmicrowave probe that makes contact withthe AUT, the aforementioned scanningopen-ended waveguide probe, an auto-matic network analyzer (more specifically,a vector network analyzer)/microwave re-ceiver, and a computer.

The mechanisms for scanning theopen-ended waveguide probe are athree-axis slide mechanism and a rota-tion mechanism that, under computercontrol, positions this probe for acquisi-tion of data at prescribed grid points ona plane very close to the AUT. This near-field scanning scheme enables captureof a maximum amount of energy radi-ated by one or multiple small antennaswhile they are DC-biased, without needfor any special fixture.

The system is controlled by user-friendly operational, data-acquisition,and data-analysis software. The dimen-sions of the near-field scan area andthe distance between grid points arespecified by the user via the computerkeyboard as inputs to a software-gener-ated control panel. After each scan, thedata-analysis software processes themeasurement data and displays the far-field radiation pattern of the AUT,

computed from the near-field measure-ments.

This work was done by Afroz Zaman,Richard Q. Lee, William G. Darby, Philip J.Barr, and Félix A. Miranda of Glenn Re-search Center; and Kevin Lambert ofAnalex Corp. Further information is containedin a TSP (see page 1).

Inquiries concerning rights for the commer-cial use of this invention should be addressed toNASA Glenn Research Center, InnovativePartnerships Office, Attn: Steve Fedor, MailStop 4–8, 21000 Brookpark Road, Cleveland,Ohio 44135. Refer to LEW-17877-1.

Probe Station and Near-Field Scanner for Testing Antennas Multiple antennas on the same substrate can be evaluated quickly and inexpensively. John H. Glenn Research Center, Cleveland, Ohio

The Open-Ended Waveguide Probe Is Scanned ina plane slightly above the AUT and is operatedin conjunction with an RF contact probe and thevector network analyzer to gather data on thenear radiation field.

LEW-17877ABPI

5-9-06 ca

Photodetector Arrays for Multicolor Visible/Infrared ImagingSeparate optical trains would not be needed for different wavelength bands.NASA’s Jet Propulsion Laboratory, Pasadena, California

Monolithic focal-plane arrays of pho-todetectors capable of imaging thesame scenes simultaneously in multiplewavelength bands in the visible and in-frared spectral regions have been pro-posed. In prior visible/infrared imag-ing systems, it has been standardpractice to use separate optical trains toform images in visible and infraredwavelength bands on separate visible-and infrared-photodetector arrays. Be-

cause the proposal would enable thedetection of images in multiple wave-length bands on the same focal plane,the proposal would make it unneces-sary to use multiple optical trains.Hence, multispectral imaging systemscould be made more compact and thedifficulties of aligning multiple opticaltrains would be eliminated.

Each pixel in an array according tothe proposal would contain stacks of sev-

eral photodetectors. The proposal is alogical extension of prior concepts of ar-rays of stacked photodetectors for imag-ing in two or three wavelength bands.For example, such an array was de-scribed in “Three-Color Focal-PlaneArray of Infrared QWIPs” (NPO-20683),NASA Tech Briefs, Vol. 24, No. 5 (May2000), page 26a.

In one proposed design, (see figure),each pixel would be divided into four


One Pixel

V/MWIRSubpixel

V/LWIRSubpixel

V Subpixel V/VLWIRSubpixel

–0.2 V

–2 V

0 V(DetectorCommon)

n+-Doped Layer

PIN Photodiode for Detecting Visibleand Near Infrared

MWIR QWIPn+-Doped Layer

n+-Doped Layer

n+-Doped Layer

n+-Doped Layer

p+-Doped Layer

Semi-Insulating GaAs Substrate

LWIR QWIP

VLWIR QWIP

Isolation Layer

Intrinsic (Undoped) Layer

Plan View of Part of Focal-Plane Array

Enlarged, Partly Schematic Cross Section of V/LWIR SubpixelShowing Electrical Connections

NPO30541

5-14-02 bs

Each Pixel of a Four-Color Focal-Plane Array would be divided into four subpixels containing stackedphotodetectors for four wavelength bands. The pixels would be identical except for the electrical con-nections for activating the detectors for different wavelength-band combinations.

subpixels, one being dedicated to a visi-ble-and-near-infrared (V) band, one to acombination of the V band and a very-long-wavelength infrared (VLWIR)band, one to a combination of the Vband and a long-wavelength infrared(LWIR) band, and one to a combinationof the V band and a medium-wavelengthinfrared (MWIR) band. For this pur-pose, each subpixel would include aGaAs-based positive/intrinsic/negative(PIN) photodiode for detection in the Vband stacked with three quantum-wellinfrared photodetectors (QWIPs), eachoptimized for one of the aforemen-tioned infrared bands. The stacks ofphotodetectors in all the subpixelswould be identical except for the electri-cal connections, which would be config-ured to activate the various wavelength-band combinations.

This work was done by Sarath Gunapala,Sumith Bandara, John Liu, and David Tingof Caltech for NASA’s Jet Propulsion Lab-oratory. Further information is contained ina TSP (see page 1).

In accordance with Public Law 96-517,the contractor has elected to retain title to thisinvention. Inquiries concerning rights for itscommercial use should be addressed to:

Innovative Technology Assets ManagementJPLMail Stop 202-2334800 Oak Grove DrivePasadena, CA 91109-8099(818) 354-2240E-mail: [email protected]

Refer to NPO-30541, volume and number of thisNASA Tech Briefs issue, and the page number.

Semiconductor bolometers that arecapable of detecting electromagnetic ra-diation over most or all of the infraredspectrum and that give background-lim-ited performance at operating tempera-tures from 20 to 300 K have been in-vented. The term “background-limitedperformance” as applied to a bolome-ter, thermopile, or other infrared detec-tor signifies that the ability to detect in-frared signals that originate outside thedetector is limited primarily by thermalnoise attributable to the background ra-diation generated external to thebolometer. The signal-to-noise ratiosand detectivities of the bolometers and

thermopiles available prior to this in-vention have been lower than thoseneeded for background-limited per-formance by factors of about 100 and10, respectively.

Like other electrically resistivebolometers, a device according to the in-vention exhibits an increase in electricalresistance when heated by infrared radi-ation. Depending on whether the deviceis operated under the customary con-stant-current or constant-voltage bias,the increase in electrical resistance canbe measured in terms of an increase involtage across the device or a decrease incurrent through the device, respectively.

In the case of a semiconductor bolome-ter, it is necessary to filter out visible andshorter-wavelength light that could in-duce photoconductivity and therebycounteract all or part of the desired in-frared-induced increase in resistance.

The basic semiconductor material of abolometer according to the invention ispreferably silicon doped with one ormore of a number of elements, each ofwhich confers a different variable tem-perature coefficient of resistance. Suit-able dopants include In, Ga, S, Se, Te, B,Al, As, P, and Sb. The concentration ofdopant preferably lies in the range be-tween 0.1 and 1,000 parts per billion.

Semiconductor Bolometers Give Background-Limited PerformanceThese devices can be fabricated inexpensively by use of established silicon-processing techniques.Ames Research Center, Moffett Field, California

The dopant and its concentration arechosen to optimize the performance ofthe bolometer, taking account of thebolometer operating temperature, thetemperature of the source of infrared ra-diation to be detected, and other rele-vant environmental factors.

An important practical advantage ofthe use of silicon, in contradistinction toother semiconductors, is that the art offabrication of electronic devices from sili-

con is mature, enabling mass productionat low cost per device. An additional ad-vantage accrues when indium is used asthe dopant: Indium can be incorporatedinto silicon over a wide range of concen-trations with little consequent change inthe basic structure of the silicon matrix.Hence, with impunity, the concentrationof indium dopant can be set at almost anydesired value in an effort to obtain thedesired electrical impedance.

This work was done by John Goebel andRobert McMurray of Ames Research Cen-ter. Further information is contained in a TSP(see page 1).

This invention has been patented by NASA(U.S. Patent No. 6,838,669). Inquiries con-cerning rights for the commercial use of this in-vention should be addressed to the Ames Tech-nology Partnerships Division at (650)604-2954. Refer to ARC-14577.


A multichannel dielectric-resonatoroscillator (DRO), built as a prototype ofa local oscillator for an X-band transmit-ter or receiver, is capable of being electri-cally tuned among and within 26 adja-cent frequency channels, each 1.16 MHzwide, in a band ranging from ≈7,040 to≈7,070 GHz. The tunability of this oscilla-tor is what sets it apart from other DROs,making it possible to use mass-producedoscillator units of identical design in di-verse X-band applications in which thereare requirements to use different fixedfrequencies or to switch among fre-quency channels.

The oscillator (see figure) includes acustom-designed voltage-controlled-oscil-lator (VCO) monolithic microwave inte-grated circuit (MMIC), a dielectric res-onator disk (“puck”), and twovaractor-coupling circuits, all laid out ona 25-mil (0.635-mm)-thick alumina sub-strate having a length and width of 17.8mm. The resonator disk has a diameter of8.89 mm and a thickness of 4.01 mm. Theoscillator is mounted in an 8.9-mm-deepcavity in a metal housing.

The VCO MMIC incorporates a nega-tive-resistance oscillator amplifier alongwith a buffer amplifier. The resonator diskis coupled to a microstrip transmissionline connected to the negative-resistanceport of the VCO MMIC. The two varactor-coupling circuits include microstrip lines,laid out orthogonally to each other, forcoupling with the resonator disk. Each var-actor microstrip line is DC-coupled to anexternal port via a microwave choke. Onevaractor is used for coarse tuning to selecta channel; the other varactor is used (1)for fine tuning across the 1.16-MHz widthof each channel and (2) as a feedback portfor a phase-lock loop. The resonator diskis positioned to obtain (1) the most desir-

Multichannel X-Band Dielectric-Resonator OscillatorUnlike other DROs, this one is electrically tunable.NASA’s Jet Propulsion Laboratory, Pasadena, California

Microstrip Lines provide coupling among the resonator disk, the tuning varactors, and the VCO MMIC.

Output

VCO MMIC

VCO MMIC

Negative-Resistance Oscillator Amplifier

Bias Voltage

50-Ω Termination

Microwave Choke

Microwave Choke

Coarse-Tuning Voltage

Fine-Tuning Voltage

Resonator Disk

SCHEMATIC DIAGRAM

GEOMETRIC LAYOUT

17.8 mm

17.8 mm

Port for Coarse-Tuning

Voltage

Port for Bias Voltage

Port for Fine-Tuning

Voltage

Resonator Disk

N


able bandwidth, (2) relatively tight cou-pling with the microstrip connected to thecoarse-tuning varactor, and (3) relativelyloose coupling with the microstrip con-nected to the fine-tuning varactor.

Measurements of performance showedthat the oscillator can be switched amongany of the 26 channels and can be phase-

locked to a nominal frequency in any chan-nel. The degree of nonlinearity of tuningwas found not to exceed 2.5 percent. Thetuning sensitivity was found to be 6.15MHz/V at a bias offset of –2 V on the phase-lock-loop varactor. The phase noise of theoscillator in free-running operation wasfound to be –107 dBc/Hz (where “dBc” sig-

nifies decibels relative to the carrier signal)at 100 kHz away from the carrier frequency.

This work was done by Narayan Mysoor,Matthew Dennis, and Brian Cook of Cal-tech for NASA’s Jet Propulsion Labora-tory. Further information is contained in aTSP (see page 1).NPO-41275

Automatic Alignment of Displacement-Measuring InterferometerCorrections are derived from fluctuations associated with circular dithering of a laser beam.NASA’s Jet Propulsion Laboratory, Pasadena, California

A control system strives to maintain thecorrect alignment of a laser beam in an in-terferometer dedicated to measuring thedisplacement or distance between two fidu-cial corner-cube reflectors. The correctalignment of the laser beam is parallel tothe line between the corner points of thecorner-cube reflectors: Any deviation fromparallelism changes the length of the opti-cal path between the reflectors, thereby in-troducing a displacement or distancemeasurement error.

On the basis of the geometrical optics ofcorner-cube reflectors, the length of theoptical path can be shown to be L = L0cosθ, where L0 is the distance between the cor-ner points and θ is the misalignment angle.Therefore, the measurement error is givenby ∆L = L0(cos θ – 1). In the usual case inwhich the misalignment is small, this errorcan be approximated as ∆L ≈ −L0θ2/2.

The control system (see figure) is imple-mented partly in hardware and partly insoftware. The control system includesthree piezoelectric actuators for rapid, fineadjustment of the direction of the laserbeam. The voltages applied to the piezo-electric actuators include components de-signed to scan the beam in a circular pat-tern so that the beam traces out a narrowcone (60 microradians wide in the initialapplication) about the direction in which itis nominally aimed. This scan is performedat a frequency (2.5 Hz in the initial applica-tion) well below the resonance frequencyof any vibration of the interferometer.

The laser beam makes a round trip toboth corner-cube reflectors and then in-terferes with the launched beam. The in-terference is detected on a photodiode.The length of the optical path is meas-ured by a heterodyne technique: A 100-kHz frequency shift between thelaunched beam and a reference beam im-poses, on the detected signal, an interfer-ometric phase shift proportional to the

length of the optical path. A phase metercomprising analog filters and specializeddigital circuitry converts the phase shiftto an indication of displacement, gener-ating a digital signal proportional to thepath length.

If the axis of the conical scan is cor-rectly aligned, then the path-length signalis steady and the path-length error re-mains constant at about −L0θ0

2/2, where,in this case, θ0 is the half cone angle. If,however, the axis of the conical scan isslightly misaligned, then the misalign-ment angle consists of a steady componentθ0 plus a small fluctuating component ∆θ.In this case, the optical-path length fluctu-ates by approximately −L0θ0∆θ. In a lock-

in amplifier, the digital path-length signalis high-pass-filtered to eliminate thesteady component, then the remainingfluctuating component is synchronouslydemodulated to generate DC signals pro-portional to the two tilt angles that char-acterize the misalignment. These signalsare superimposed upon the voltages ap-plied to the piezoelectric actuators tocounteract the misalignment.

This work was done by Peter Halverson,Martin Regehr, Robert Spero, Oscar Alvarez-Salazar, Frank Loya, and Jennifer Logan ofCaltech for NASA’s Jet Propulsion Labora-tory. Further information is contained in aTSP (see page 1).NPO-40957

Phase Meter High-Voltage

Amplifier

Control Law Block High-Pass Filter

Sine-Wave Generator

Digital Lock-In Amplifier

Corner-Cube Reflector Corner-Cube

Reflector

Laser-Beam-Launching Optics and Piezoelectric Actuators

Auxiliary Laser

Quad Diode Photodetectors

Digital-to- Analog

Converter

NPO40957ABPI

6-7-04 bs

+Conversion Matrix for

Piezoelectric- Actuator Voltages

This Simplified Schematic Diagram of the control system illustrates the control of the tilt of the laserbeam about one of two orthogonal tilt axes.


Software

Earth Observing SystemData Gateway

The Earth Observing System DataGateway (EDG) software provides a“one-stop-shopping” standard interfacefor exploring and ordering Earth-sci-ence data stored at geographically dis-tributed sites. EDG enables a user to dothe following:• Search for data according to high-level

criteria (e.g., geographic location,time, or satellite that acquired thedata);

• Browse the results of a search, viewingthumbnail sketches of data that satisfythe user’s criteria; and

• Order selected data for delivery to aspecified address on a chosen medium(e.g., compact disk or magnetic tape).EDG consists of (1) a component

that implements a high-levelclient/server protocol, and (2) a col-lection of C-language libraries that im-plement the passing of protocol mes-sages between an EDG client and oneor more EDG servers. EDG servers arelocated at sites usually called “Distrib-uted Active Archive Centers” (DAACs).Each DAAC may allow access to manyindividual data items, called “granules”(e.g., single Landsat images). Relatedgranules are grouped into collectionscalled “data sets.” EDG enables a userto send a search query to multipleDAACs simultaneously, inspect the re-sulting information, select browseablegranules, and then order selected datafrom the different sites in a seamlessfashion.

This program was developed by Robin Pfis-ter of Goddard Space Flight Center andJoe McMahon, James Amrhein, Ed Sefert,Lorena Marsans, Mark Solomon, and MarkNestler of Global Science & Technology, Inc.Further information is contained in a TSP(see page 1).GSC-14938-1

Power User InterfacePower User Interface 5.0 (PUI) is a

system of middleware that is an alterna-tive to the computer program describedin the immediately preceding article.Written for expert users in the Earth-sci-ence community, PUI enables expeditedordering of data granules on the basis ofspecific granule-identifying information

that the users already know or can as-semble. PUI also enables expert users toperform quick searches for orderable-granule information for use in prepar-ing orders. PUI 5.0 is available in twoversions (note: PUI 6.0 has command-line mode only): a Web-based applica-tion program and a UNIX command-line-mode client program. Both versionsinclude modules that perform data-granule-ordering functions in conjunc-tion with external systems. The Web-based version works with EarthObserving System Clearing House(ECHO) metadata catalog and order-entry services and with an open-sourceorder-service broker server component,called the Mercury Shopping Cart, thatis provided separately by Oak Ridge Na-tional Laboratory through the Depart-ment of Energy. The command-line ver-sion works with the ECHO metadata andorder-entry process service. Both ver-sions of PUI ultimately use ECHO toprocess an order to be sent to a dataprovider. Ordered data are providedthrough means outside the PUI softwaresystem.

This program was developed by Robin Pfis-ter of Goddard Space Flight Center andJoe McMahon of Global Science & Technol-ogy, Inc. Further information is contained ina TSP (see page 1).GSC-14939-1

Mercury Shopping Cart Interface

Mercury Shopping Cart Interface(MSCI) is a reusable component of thePower User Interface 5.0 (PUI) pro-gram described in the immediately pre-ceding article. MSCI is a means of en-capsulating the logic and informationneeded to describe an orderable itemconsistent with Mercury Shopping Cartservice protocol. Designed to be usedwith Web-browser software, MSCI gener-ates Hypertext Markup Language(HTML) pages on which ordering in-formation can be entered. MSCI com-prises two types of Practical Extractionand Report Language (PERL) modules:template modules and shopping-cartlogic modules. Template modules gen-erate HTML pages for entering the re-quired ordering details and enable sub-mission of the order via a HypertextTransfer Protocol (HTTP) post. Shop-

ping-cart modules encapsulate the logicand data needed to describe an individ-ual orderable item to the Mercury Shop-ping Cart service. These modules evalu-ate information entered by the user todetermine whether it is sufficient forthe Shopping Cart service to process theorder. Once an order has been passedfrom MSCI to a deployed MercuryShopping Cart server, there is no fur-ther interaction with the user.

This program was developed by Robin Pfis-ter of Goddard Space Flight Center andJoe McMahon of Global Science & Technol-ogy, Inc. Further information is contained ina TSP (see page 1).GSC-14940-1

Cassini Archive TrackingSystem

The Cassini Archive Tracking System(CATS) is a computer program that en-ables tracking of scientific data trans-fers from originators to the PlanetaryData System (PDS) archives. WithoutCATS, there is no systematic means oflocating products in the archive processor ensuring their completeness. Bykeeping a database of transfer commu-nications and status, CATS enables theCassini Project and the PDS to effi-ciently and accurately report on archivestatus. More importantly, problem areasare easily identified through cus-tomized reports that can be generatedon the fly from any Web-enabled com-puter. A Web-browser interface andclearly defined authorization schemeprovide safe distributed access to thesystem, where users can perform func-tions such as create customized reports,record a transfer, and respond to atransfer. CATS ensures that Cassini pro-vides complete science archives to thePDS on schedule and that thosearchives are available to the sciencecommunity by the PDS. The three-tierarchitecture is loosely coupled and de-signed for simple adaptation to multi-mission use. Written in the Java pro-gramming language, it is portable andcan be run on any Java-enabled Webserver.

This work was done by Diane Conner,Elias Sayfi, and Adrian Tinio of Caltech forNASA’s Jet Propulsion Laboratory. Fur-ther information is contained in a TSP (seepage 1).


This software is available for commerciallicensing. Please contact Karina Edmonds ofthe California Institute of Technology at(818) 393-2827. Refer to NPO-40951.

Architecture Adaptive Computing Environment

Architecture Adaptive ComputingEnvironment (aCe) is a software systemthat includes a language, compiler, andrun-time library for parallel computing.aCe was developed to enable program-mers to write programs, more easilythan was previously possible, for a vari-ety of parallel computing architectures.Heretofore, it has been perceived to bedifficult to write parallel programs forparallel computers and more difficult toport the programs to different parallelcomputing architectures. In contrast,aCe is supportable on all high-perform-ance computing architectures. Cur-rently, it is supported on LINUX clus-ters. aCe uses parallel programmingconstructs that facilitate writing of paral-lel programs. Such constructs were usedin single-instruction/multiple-data(SIMD) programming languages of the1980s, including Parallel Pascal, ParallelForth, C*, *LISP, and MasPar MPL. InaCe, these constructs are extended and

implemented for both SIMD and multi-ple-instruction/multiple-data (MIMD)architectures. Two new constructs incor-porated in aCe are those of (1) scalarand virtual variables and (2) pre-com-puted paths. The scalar-and-virtual-vari-ables construct increases flexibility inoptimizing memory utilization in vari-ous architectures. The pre-computed-paths construct enables the compiler topre-compute part of a communicationoperation once, rather than computingit every time the communication opera-tion is performed.

This program was written by John E. Dor-band of Goddard Space Flight Center.For further information, contact the GoddardInnovative Partnerships Office at (301) 286-5810.GSC-14911-1

Computing Fault Displacements From Surface Deformations

Simplex is a computer program thatcalculates locations and displacementsof subterranean faults from data onEarth-surface deformations. The calcu-lation involves inversion of a forwardmodel (given a point source represent-ing a fault, a forward model calculates

the surface deformations) for displace-ments, and strains caused by a fault lo-cated in isotropic, elastic half-space. Theinversion involves the use of nonlinear,multiparameter estimation techniques.The input surface-deformation data canbe in multiple formats, with absolute ordifferential positioning. The input datacan be derived from multiple sources,including interferometric synthetic-aperture radar, the Global PositioningSystem, and strain meters. Parameterscan be constrained or free. Estimatescan be calculated for single or multiplefaults. Estimates of parameters are ac-companied by reports of their covari-ances and uncertainties. Simplex hasbeen tested extensively against forwardmodels and against other means of in-verting geodetic data and seismic obser-vations.

This work was done by Gregory Lyzenga,Jay Parker, and Andrea Donnellan of Caltechand Wendy Panero of Ohio State Universityfor NASA’s Jet Propulsion Laboratory.Further information is contained in a TSP(see page 1).

This software is available for commerciallicensing. Please contact Karina Edmonds ofthe California Institute of Technology at(818) 393-2827. Refer to NPO-41078.


Materials

Oxygen-Permeable, Hydrophobic Membranes of Silanized α-Al2O3These membranes perform better than do organic polymer oxygen-diffusion membranes.Lyndon B. Johnson Space Center, Houston, Texas

Membranes made of silanized alu-mina have been prepared and tested asprototypes of derivatized ceramic mem-branes that are both highly permeableto oxygen and hydrophobic. Improvedoxygen-permeable, hydrophobic mem-branes would be attractive for use in sev-eral technological disciplines, includingsupporting high-temperature aqueous-phase oxidation in industrial productionof chemicals, oxygenation of aqueousstreams for bioreactors, and oxygena-tion of blood during open-heart surgeryand in cases of extreme pulmonary dys-function. In comparison with organicpolymeric oxygen-permeable mem-branes now commercially available, thederivatized ceramic membranes aremore chemically robust, are capable ofwithstanding higher temperatures, andexhibit higher oxygen-diffusion coeffi-cients.

Membranes made from alumina aswell as such other ceramics as titania andzirconia are permeable to oxygen andcapable of withstanding higher tempera-tures. However, without modification,these ceramics are also hydrophilic.Hence, it is necessary to modify the sur-face properties of these ceramics to ren-

der them hydrophobic. For a series ofexperiments, the prototype membraneswere made from α-Al2O3 with pore sizesfrom 5 to 200 nm. Hydrophobic molecu-lar groups were attached to each α-Al2O3

membrane through silanization, using asuitable trimethoxy- or triethoxysilane(see figure).

In the experiments, both the silanizedα-Al2O3 membranes and an organicpolymer membrane based on polydi-methylsiloxane (PDMS) were used asmedia for the transport of oxygen froma constant-pressure gas phase into a re-circulating aqueous stream. Coefficientsof diffusion of O2 and H2O across themembranes were measured. At roomtemperature, the silanized α-Al2O3 mem-branes exhibited oxygen-diffusion coef-

ficients ranging from 1.24 to 5.75 timesthat of the PDMS membrane, the valuein each case depending on the pore sizeand on which hydrophobic functionalgroups were present. Water-loss rates ofthe silanized α-Al2O3 membranes werefound to be as much as two orders ofmagnitude below that of the PDMSmembrane. In one test at a temperatureof 90 °C, one of the silanized α-Al2O3

membranes exhibited an oxygen-diffu-sion coefficient 23.9 times that of thePDMS membrane at 23 °C.

This work was done by James E. Atwaterand James R. Akse of Umpqua Research Co.for Johnson Space Center. For further in-formation, contact the Johnson InnovativePartnerships Office at (281) 483-3809.MSC-23384

OH OH OH OH

Al2O3 Surface

(CH3O)3Si––R

or(C2H5O)3Si––R

O O O O

R R R R

O O O OSi Si Si Si

MSC23384

1-10-02 bs

Hydrophobic Groups (R) have been attached to alumina surfaces by silanization. Thus far, four hy-drophobic groups have been studied: R = —CH2—CH2—CH2—CH3, R = —CH2—CH2(CF2)7CF3, R = —(CH2)11CH3, and R = —(CH2)17CH3.

Turbine inlet guide vanes have beenfabricated from composites of siliconcarbide fibers in silicon carbide matri-ces. A unique design for a cloth madefrom SiC fibers makes it possible to real-ize the geometric features necessary toform these vanes in the same airfoilshapes as those of prior metal vanes.

The fiber component of each of thesevanes was made from SiC-fiber clothcoated with boron nitride. The matrixwas formed by chemical-vapor infiltra-

tion with SiC, then slurry-casting of SiC,followed by melt infiltration with silicon.

These SiC/SiC vanes were found to becapable of withstanding temperatures400 °F (222 °C) greater than those thatcan be withstood by nickel-base-superal-loy turbine airfoils now in common usein gas turbine engines. The higher tem-perature capability of SiC/SiC parts isexpected to make it possible to use themwith significantly less cooling than isused for metallic parts, thereby enabling

engines to operate more efficientlywhile emitting smaller amounts of NOx

and CO. The SiC/SiC composite vanes were fab-

ricated in two different configurations.Each vane of one of the configurationshas two internal cavities formed by a webbetween the suction and the pressuresides of the vane. Each vane of the otherconfiguration has no web (see Figure 1).

It is difficult to fabricate componentshaving small radii, like those of the trail-

SiC Composite Turbine Vanes Y-cloth was conceived to provide fiber reinforcement for sharp trailing edges. John H. Glenn Research Center, Cleveland, Ohio


ing edges of these vanes, by use of stiffstoichiometric SiC fibers currently pre-ferred for SiC/SiC composites. To satisfythe severe geometric and structural re-quirements for these vanes, the afore-mentioned unique cloth design, de-noted by the term “Y-cloth,” was

conceived (see Figure 2). In the regionsaway from the trailing edge, the Y-clothfeatures a fiber architecture that hadbeen well characterized and successfullydemonstrated in combustor liners. Toform a sharp trailing edge (having a ra-dius of 0.3 mm), the cloth was split intotwo planes during the weaving process.The fiber tows forming the trailing-edgesection were interlocked, thereby en-hancing through-thickness strength ofthe resulting composite material.

For vanes of the webless configuration,each made from a layup of six plies of Y-cloth, the length of each Y-cloth layer wascut so that the two strips correspondingto the aforementioned two planes wouldwrap around the perimeter of a graphitevane preform tool with a 10-mm overlap.The overlap was used to join the twostrips in a fringe splice. To make the ex-ternal sixth ply, a standard woven clothwas cut to the required final length and afringe splice joined the two ends of the

cloth at the trailing edge. The cloth wasthen prepregged. The entire assemblywas then placed into an aluminum com-paction tool designed to form the outernet shape of the vane. After the prepregmaterial was allowed to dry, the preformwas removed from the aluminum toolingand placed into an external graphite toolbefore being shipped to a vendor for ma-trix infiltration.

To make the SiC fiber preform for avane having an internal web, a slightlydifferent initial approach was followed.Each of two sections forming the inter-nal cavities (and ultimately the web) wascreated by first slipping two concentriclayers of a two-dimensional, 2-by-2, ±45°-braided tube around a net-shapegraphite mandrel. The tubes on bothmandrels were prepregged and allowedto dry. The resulting two subassemblieswere put together, then four additionalplies were wrapped around them in thesame fashion as that described above forthe six plies of the vaneless configuration.

The consolidation of the SiC fiberpreforms into SiC/SiC composite partswas performed by commercial vendorsusing their standard processes. The ca-pability of two of the webless SiC/SiCturbine vanes was demonstrated in testsin a turbine environment. The tests in-cluded 50 hours of steady-state opera-tion and 102 two-minute thermal cycles.A surface temperature of 1,320 °C wasreached during the tests.

This work was done by Anthony M.Calomino and Michael J. Verrilli of GlennResearch Center. Further information iscontained in a TSP (see page 1).

Inquiries concerning rights for the commer-cial use of this invention should be addressedto NASA Glenn Research Center, InnovativePartnerships Office, Attn: Steve Fedor, MailStop 4–8, 21000 Brookpark Road, Cleve-land, Ohio 44135. Refer to LEW-17882-1.

Figure 1. SiC/SiC Composite Turbine Vanes were fabricated in two configurations: one webless, onewith an internal web.

50 mm50 mm

LEW-17882-1 FIG1 ABPI

06-23-05 LE

LEW-17882-1 FIG2 ABPI

06-23-05 LE

Figure 2. Y-Cloth made it possible to form trail-ing edges to the required small radius.


Mechanics

A device denoted as a bumper assem-bly for a spacecraft payload containercomprises an interior structure sur-rounded by skin or some other protectiveenclosure (see figure). When arrangedwith three or more like assemblies, thisbumper assembly is designed to securethe interior structure within a payload’sprotective enclosure during the stressesendured in flight and, if required, recov-ery of the payload. Furthermore, properuse of this innovation facilitates the abil-ity of designers and engineers to maxi-mize the total placement area for compo-nents, thus increasing utilization of veryvaluable and limited space.

Typically, the interior structure in-cludes substantially circular decks nomi-nally orthogonal to and centered on thecylindrical axis with the decks connectedby axial columns. At one end of the cylin-der, the interior structure is attached tothe skin by use of fasteners. At other lo-cations, the column portions of the inte-rior structure are connected to the cylin-drical skin via four bumper assemblies.The bumper assemblies provide lateral,(that is, radial and circumferential) sup-port while allowing sliding parallel to thecylindrical axis to accommodate axial ex-pansion and contraction. The shape ofthe interior structure can be varied fromthe stated “typical” one. The attachmentof the end of any interior structure to aprotective enclosure, while placing thebumper assemblies in a radial symmetricpattern on the structure’s other end,

would allow similar support during theprocess of positioning and securing thebumper assemblies.

Each bumper assembly includes twomating wedges held together by a bolt.The bolt is inserted through a clearancehole in one wedge to engage a threadedhole in the other wedge. The position-ing and securing of the interior struc-ture can be adjusted by turning the boltto slide the wedges along their matingsloped surfaces. This arrangement ofthe interior structure is accomplished

from the structure’s outside area anddoes not require access holes or sur-faces machined within its protective en-closure to achieve that accessibility. Thisaccessibility minimizes the time neededto finish the securing of the interiorstructure within a payload’s protectiveenclosure.

This work was done by Orville N. Fleming,Jr., of Northrop Grumman Corp. for God-dard Space Flight Center. Further infor-mation is contained in a TSP (see page 1).

Retaining Device for the Interior Structure of a Spacecraft PayloadDevice protects without penalizing interior space.Goddard Space Flight Center, Greenbelt, Maryland

Tool for Torquing Circular Electrical-Connector CollarsA simple tool exerts a strong grip.Goddard Space Flight Center, Greenbelt, Maryland

An improved tool has been devised forapplying torque to lock and unlockknurled collars on circular electrical con-nectors. The tool was originally designedfor, and used by, astronauts working inouter space on the Hubble Space Tele-scope (HST). The tool is readily adapt-

able to terrestrial use in installing and re-moving the same or similar circular elec-trical connectors as well as a wide varietyof other cylindrical objects, the tighteningand loosening of which entail consider-able amounts of torque.

Other tools developed previously for

mating or de-mating electrical connec-tor collars were either designed for useon specific connectors or too genericand incapable of applying the requisiteamount of torque [40 lb-in. (4.52 N-m)]for the HST application. In contrast, thepresent improved tool can be used on a

Payload Skin (Transparent for Clarity )

Bumper Assembly

Payload Interior Structure

Bumper Assembly

Payload Interior Structure

Top Taper Part

Fastener Bottom Taper Part

INTERIOR STRUCTURE WITHIN ITS SKIN

INTERIOR STRUCTURE ASSEMBLY— CLOSE-UP

BUMPER ASSEMBLY— MINIMUM ADJUSTMENT

GSC-14991-1

ABPI

11-3-05 CA

The Bumper Assembly for a spacecraft payload container protects the interior of the payload container,as shown in different magnified views.


variety of connector sizes and is capableof applying the requisite amount oftorque. Indeed, only a moderateamount of hand clasping force [25 lb(≈111 N)] is necessary for applying dou-ble the requisite amount of torque.

The tool consists of two stainless steelarms that pivot about a common point.Attached to the gripping jaws on thearms are a total of four flat pads, made ofcommercially available rubberlike epoxy.The pads make tangential contact withthe circular connector collar. Under thegripping force, the pads deform intogreater conformity with the gripped ob-

ject and are thereby capable of exertinga greater tangential frictional force.Hence, this jaw-and-pad combination en-ables the tool to fit circular connectors ofdifferent diameters and to exert greatertorque than could otherwise be applied.A simple spring-lever resists the user’shand-grasping force with just enoughforce to return the gripping jaws to thewide-open position.

Although deformation of the pads inrepeated use of the tool degrades per-formance, the amount of degradationmay be acceptable in some applicationsand was acceptable in the original HST

application. In that application, the toolperformed as required when used toloosen, then later to tighten, 36 connec-tors in an operation to remove and re-place a power-control unit. Theoreti-cally, the tool could be used to performthe operation a total of five times.

This work was done by Kathryn Gaulkeand Russel Werneth of Goddard SpaceFlight Center, John Grunsfeld of JohnsonSpace Center, and Patrick O’Neill and RussSnyder of Swales Aerospace. Further informa-tion is contained in a TSP (see page 1).GSC-14670-1


Machinery/Automation

A system for continuous, rapid deaer-ation of hydraulic oil has been built toreplace a prior system that effecteddeaeration more slowly in a cyclic pres-sure/vacuum process. Such systems areneeded because (1) hydraulic oil has anaffinity for air, typically containing be-tween 10 and 15 volume percent of airand (2) in the original application forwhich these systems were built, there is arequirement to keep the proportion ofdissolved air below 1 volume percent be-cause a greater proportion can lead topump cavitation and excessive softnessin hydraulic-actuator force-versus-dis-placement characteristics. In addition toovercoming several deficiencies of theprior deaeration system, the present sys-tem removes water from the oil.

The system (see figure) includes apump that continuously circulates oil

at a rate of 10 gal/min (38 L/min) be-tween an 80-gal (303-L) airless reser-voir and a tank containing a vacuum.When the circulation pump is started,oil is pumped, at a pressure of 120 psi(827 kPa), through a venturi tubebelow the tank with a connection to astandpipe in the tank. This actiondraws oil out of the tank via the stand-pipe. At the same time, oil is sprayedinto the tank in a fine mist, thereby ex-posing a large amount of oil to the vac-uum. When the oil level in the tankfalls below the lower of two levelswitches, a vacuum pump is started,drawing a hard vacuum on the tankthrough a trap that collects any oil andwater entrained in the airflow. Whenthe oil level rises above higher of thetwo level switches or when the system isshut down, a solenoid valve between

the tank and the vacuum pump isclosed to prevent suction of oil into thevacuum pump.

Critical requirements that the systemis designed to satisfy include the follow-ing:• The circulation pump must have

sufficient volume and pressure tooperate the venturi tube and spraynozzles.

• The venturi tube must be sized to empty the tank (except for the oil retained by the standpipe) andmaintain a vacuum against the vac-uum pump.

• The tank must be strong enough towithstand atmospheric pressure againstthe vacuum inside and must have suffi-cient volume to enable exposure of asufficiently large amount of sprayed oilto the vacuum.

System for Continuous Deaeration of Hydraulic OilThe proportion of dissolved air is reliably maintained below 1 volume percent. John F. Kennedy Space Center, Florida

Oil Is Circulated continuously between the reservoir and the tank. Oil is sprayed in the tank to maximize its exposure to vacuum for rapid removal of dis-solved air and water.

Airless Reservoir

Float

Vacuum Gauge

Solenoid Valve

Vacuum Transducer

Pressure Transducer

Trap

Vacuum Pump

Circulation Pump

Venturi Tube

Oil

Oil

Tank

Level Switches

Oil Sprays

Standpipe

KSC-12528 ABPI

6-27-05 es

The apparatus would include a solarphotovoltaic panel mounted on theroof and a panellike assemblymounted in a window opening. The

window-mounted assembly (see fig-ure) would include a stack of thermo-electric devices sandwiched betweentwo heat sinks. A fan would circulate

interior air over one heat sink. An-other fan would circulate exterior airover the other heat sink. The fans andthe thermoelectric devices would bepowered by the solar photovoltaicpanel. By means of a double-pole, dou-ble-throw switch, the panel voltage fedto the thermoelectric stack would beset to the desired polarity: For coolingoperation, the chosen polarity wouldbe one in which the thermoelectric de-vices transport heat from the insideheat sink to the outside one; for heat-ing operation, the opposite polaritywould be chosen.

Because thermoelectric devices aremore efficient in heating than in cool-ing, this apparatus would be more effec-tive as a heater than as a cooler. How-ever, if the apparatus were to includemeans to circulate air between the out-side and the inside without opening thewindows, then its effectiveness as acooler in a hot, sunny location wouldbe increased.

This work was done by Richard T.Howard of Marshall Space Flight Center.Further information is contained in a TSP(see page 1).

This invention has been patented by NASA(U.S. Patent No.6,662,572). Inquiries concern-ing nonexclusive or exclusive license for its com-mercial development should be addressed toSammy Nabors, MSFC Commercialization As-sistance Lead, at [email protected] to MFS-31751-1.


• The spray nozzles must be sized to at-omize the oil and to ensure that therate of flow of sprayed oil does not ex-ceed the rate at which the venturi ac-tion can empty the tank.

• The vacuum pump must produce ahard vacuum against the venturi tubeand continue to work when it ingestssome oil and water.

• Fittings must be made vacuum tight(by use of O-rings) to prevent leakageof air into the system. The system is fully automatic, and

can be allowed to remain in operationwith very little monitoring. It is capa-ble of reducing the air content of theoil from 11 to less than 1 volume per-cent in about 4 hours and to keep the

water content below 100 parts per mil-lion.

This work was done by Christopher W.Anderson of Lockheed Martin Space Opera-tions for Kennedy Space Center. Furtherinformation is contained in a TSP (see page1).KSC-12528

Solar-Powered Cooler and Heater for an Automobile InteriorThermoelectric devices and fans would run on solar power.Marshall Space Flight Center, Alabama

An Assembly Mounted in a Window Opening of an automobile would include thermoelectric devicesthat would transfer heat between interior and exterior circulating airflows. The thermoelectric devicesand the fans in the assembly would be powered by a solar photovoltaic panel mounted on the roof.

Fans

Inside Outside

Heat Sinks

Flow of Outside Air

Panel for Mounting in

Window Opening

Flow of Inside Air Air Ducts

Stack of Thermoelectric

Devices

MFS-31751-1 ABPI

5-3-05 bs


Manufacturing & Prototyping

An improved method has been devisedfor using directed, hyperthermal beamsof oxygen atoms and ions to impart de-sired textures to the tips of polymethyl-methacrylate [PMMA] optical fibers to beused in monitoring the glucose contentof blood. The improved method incorpo-rates, but goes beyond, the method de-scribed in “Texturing Blood-Glucose-Monitoring Optics Using Oxygen Beams”(LEW-17642-1), NASA Tech Briefs, Vol. 29,No. 4 (April 2005), page 11a.

The basic principle of operation ofsuch a glucose-monitoring sensor is as fol-lows: The textured surface of the opticalfiber is coated with chemicals that inter-act with glucose in such a manner as tochange the reflectance of the surface.Light is sent down the optical fiber and isreflected from, the textured surface. Theresulting change in reflectance of thelight is measured as an indication of theconcentration of glucose.

The required texture on the ends ofthe optical fibers is a landscape of micro-scopic cones or pillars having high aspectratios (microscopic structures being tallerthan they are wide). The average distancebetween hills must be no more than about5 µm so that blood cells (which are wider)cannot enter the valleys between the hills,where they would interfere with opticalsensing of glucose in the blood plasma.On the other hand, the plasma is re-quired to enter the valleys, and high as-pect ratio structures are needed to maxi-mize the surface area in contact with theplasma, thereby making it possible to ob-tain a given level of optical glucose-meas-urement sensitivity with a relatively small

volume of blood. There is an additionalrequirement that the hills be wideenough that a sufficient amount of lightcan propagate into them and, after reflec-tion, can propagate out of them.

The method described in the citedprior article produces a texture compris-ing cones and pillars that conform to theaverage-distance and aspect-ratio require-ments. However, a significant fraction ofthe cones and pillars are so narrow thatnot enough light can propagate alongthem. The improved method makes itpossible to form wider cones and pillarswhile still satisfying the average-distanceand aspect-ratio requirements.

In the improved method, as in the previ-ously reported method, multiple opticalfibers are first bundled together for simulta-neous texturing of their distal tips. How-ever, prior to texturing by exposure to anoxygen beam, the tips are first coated byvapor deposition of a thin, sparse layer ofaluminum: The exposure to the aluminumvapor must be short enough (typically ofthe order of seconds) so that the aluminumnucleates into islands separated by un-coated areas. The coated tips are texturedby exposure to a directed beam of hyper-thermal (kinetic energy >1 eV) oxygenatoms and/or ions in a vacuum chamber, asin the previously reported method. Thealuminum islands partially shield the un-derlying PMMA from oxidation and ero-sion by the beam, so that the cones or pil-lars remaining after texturing are widerthan they would otherwise be. To some ex-tent, the dimensions of the hills and the dis-tances between them can be tailoredthrough choice of the thickness of the alu-

minum coat and/or the oxygen-beam flu-ence. The figure illustrates an example oftexturing of the tip of a PMMA optical fiberwithout and with prior aluminum coating.

This work was done by Bruce A. Banks ofGlenn Research Center. Further informa-tion is contained in a TSP (see page 1).

Inquiries concerning rights for the commer-cial use of this invention should be addressedto NASA Glenn Research Center, InnovativePartnerships Office, Attn: Steve Fedor, MailStop 4–8, 21000 Brookpark Road, Cleveland,Ohio 44135. Refer to LEW-17975-1.

Improved Oxygen-Beam Texturing of Glucose-Monitoring Optics Textures can be more nearly optimized for greater utilization of light. John H. Glenn Research Center, Cleveland, Ohio

These Scanning Electron Micrographs show theresults of oxygen-beam texturing of the tips oftwo PMMA optical fibers: one that was notcoated and one that was sparsely coated withaluminum.

WITH ALUMINUM COATING

WITHOUT ALUMINUM COATING 10µm 1

Tool for Two Types of Friction Stir WeldingThe same mechanism could be used for conventional or self-reacting FSW.Marshall Space Flight Center, Alabama

A tool that would be useable in bothconventional and self-reacting frictionstir welding (FSW) has been proposed.The tool would embody both a prior

tooling concept for self-reacting FSWand an auto-adjustable pin-tool (APT)capability developed previously as anaugmentation for conventional FSW.

Some definitions of terms are prereq-uisite to a meaningful description of theproposed tool. In conventional FSW, de-picted in Figure 1, one uses a tool that

includes (1) a rotating shoulder on top(or front) of the workpiece and (2) a ro-tating pin that protrudes from the shoul-der into the depth of the workpiece. Themain axial force exerted by the tool onthe workpiece is reacted through aridged backing anvil under (behind) theworkpiece. When conventional FSW isaugmented with an APT capability, thedepth of penetration of the pin into theworkpiece is varied in real time by a po-sition- or force-control system that ex-tends or retracts the pin as needed to ob-tain the desired effect.

In self-reacting (also known as self-re-acted) friction stir welding (SR-FSW),there are two rotating shoulders: one ontop (or front) and one on the bottom(or back) of the workpiece. In this case,

a threaded shaft protrudes from the tipof the pin to beyond the back surface ofthe workpiece. The back shoulder isheld axially in place against tension by anut on the threaded shaft. The mainaxial force exerted on the workpiece bythe tool and front shoulder is reactedthrough the back shoulder and thethreaded shaft, back into the FSW ma-chine head, so that a backing anvil is nolonger needed. A key transmits torquebetween the bottom shoulder and thethreaded shaft, so that the bottom shoul-der rotates with the shaft. A tool for SR-FSW embodying this concept was re-ported in “Mechanism for Self-ReactedFriction Stir Welding” (MFS-31914),NASA Tech Briefs, Vol. 28, No. 10 (Octo-ber 2004), page 53.

In its outward appearance, the pro-posed tool (see Figure 2) would fit theabove description of an SR-FSW tool. Inthis case, the FSW machine would havean APT capability and the pin would bemodified to accept a bottom shoulder.The APT capability could be used to varythe distance between the front and backshoulders in real time to accommodateprocess and workpiece-thickness varia-tions. The tool could readily be con-verted to a conventional FSW tool, withor without APT capability, by simply re-placing the modified pin with a conven-tional FSW pin.

This work was done by Robert Carter ofMarshall Space Flight Center. Further in-formation is contained in a TSP (see page 1).

This invention has been patented by NASA(U.S. Patent No. 6,758,382). Inquiries con-cerning nonexclusive or exclusive license for itscommercial development should be addressed toSammy Nabors, MSFC Commercialization As-sistance Lead, at [email protected] to MFS-31647-1.


Figure 1. In Conventional FSW, the force exerted by the tool on the workpiece is reacted by the backing anvil.

2.5

300 to 400 rpm

Pin

Shoulder

Backing Anvil

Workpiece

Travel 4 to 5 Inches per Minute

MFS-31647-1 Fig 1 ABPI

4-2-05 es

MFS-31647-1 Fig 2 ABPI

1-25-06 CC

Travel Rotation

Pinch Force

Workpiece

Front Shoulder

Back Shoulder

Figure 2. The Proposed Tool would include a pinmodified to accept a back shoulder.


Bio-Medical

A proposed compact, portable instru-ment would sample micron-sized air-borne particles, would discriminate be-tween biological ones (e.g., bacteria)and nonbiological ones (e.g., dust parti-cles), and would collect the detected bi-

ological particles for further analysis.The instrument is intended to satisfy agrowing need for means of rapid, inex-pensive collection of bioaerosols in a va-riety of indoor and outdoor settings.Purposes that could be served by such

collection include detecting airbornepathogens inside buildings and theirventilation systems, measuring concen-trations of airborne biological contami-nants around municipal waste-process-ing facilities, monitoring airborne

Instrument Would Detect and Collect Biological AerosolsSamples would be quickly collected on substrates that would be analyzed automatically.Marshall Space Flight Center, Alabama

A proposed apparatus would applycontrolled cyclic forces to both feet forthe purpose of preventing the loss ofbone density in a human subject whosebones are not subjected daily to the me-chanical loads of normal activity in nor-mal Earth gravitation. The apparatus wasconceived for use by astronauts on longmissions in outer space; it could also beused by bedridden patients on Earth, in-cluding patients too weak to generate thenecessary forces by their own efforts.

The apparatus (see figure) would be amodified version of a bicyclelike exer-cise machine, called the cycle ergometerwith vibration isolation system (CEVIS),now aboard the International Space Sta-tion. Attached to each CEVIS pedalwould be a computer-controlled stress/vibration exciter connected to the heelportion of a special-purpose pedal. Theuser would wear custom shoes thatwould amount to standard bicycle shoesequipped with cleats for secure attach-ment of the balls of the feet to the spe-cial-purpose pedals.

If possible, prior to use of the appara-tus, the human subject would wear aportable network of recording accelerom-eters, while walking, jogging, and run-ning. The information thus gatheredwould be fed to the computer, wherein itwould be used to make the exciters applyforces and vibrations closely approximat-ing the forces and vibrations experiencedby that individual during normal exercise.It is anticipated that like the forces ap-plied to bones during natural exercise,

these artificial forces would stimulatethe production of osteoblasts (bone-forming cells), as needed to prevent orretard loss of bone mass.

In addition to helping to prevent dete-rioration of bones, the apparatus couldbe used in treating a person already suf-fering from osteoporosis. For this pur-pose, the magnitude of the applied forcescould be reduced, if necessary, to a level

at which weak hip and leg bones wouldstill be stimulated to produce osteoblastswithout exposing them to the full stressesof walking and thereby risking fracture.

This work was done by Jessica Hauss, JohnWood, Jason Budinoff, and Michael Correiaof Goddard Space Flight Center andRudolf Albrecht of ESA. Further informationis contained in a TSP (see page 1).GSC-14700-1

Stationary Apparatus Would Apply Forces of Walking to FeetThe forces would be tailored to prevent loss of bone density.Goddard Space Flight Center, Greenbelt, Maryland

An Exercise Machine would include computer-controlled stress-vibration exciters that would applyforces substituting for normal exercise forces to stimulate bones.

Handlebar withComputerControls

Shoulder Padand

Restraint

BungeeRetention

Strap

Stress/VibrationExciter on Crank

Shoe andPedal

GSC-14700-1 ABPI

10-23-03 es


effluents from suspected biowarfare fa-cilities, and warning of the presence ofairborne biowarfare agents.

The instrument would be based partlyon a conventional aerosol-particlecounter and partly on a fluorescence sub-system for identifying biological particles.Aerosol particles would be drawnthrough a series of aerodynamic lenses(nozzles sized and shaped to focus vari-ously sized particles into a narrowstream). The lenses would be designed sothat only respirable particles would endup in a narrow outlet stream flowingacross the optical path of a pulsed ultravi-olet laser in the fluorescence-based sub-system. Before reaching the optical path

of the pulsed ultraviolet laser, the aerosolparticles would cross the beam of a con-tinuous-wave semiconductor diode laserthat would be used to size the particles. Ifthe size of an individual particle wasfound to be within a certain range, the ul-traviolet laser would be triggered to fireas the particle crossed its path, therebydramatically reducing power require-ments for autonomous operation.

The pulse of ultraviolet light wouldexcite fluorescence in the particle. Thefluorescent light would be collected andsplit into three separate spectral bandsby use of lenses, dichroic filters, andband-pass filters. The outputs of pho-todetectors for the three spectral bands

would be processed to determinewhether the particle could be of biologi-cal origin. The indication of a possiblebiological particle would cause anaerosol-sampling module to be turnedon to collect particles on a solid sub-strate. The substrate would be placedunder an automated microscopeequipped with a video camera, the out-put of which would be digitized andprocessed by image-analysis software toidentify the collected particles.

This work was done by Steve Savoy andMike Mayo of Nanohmics, Inc. for Mar-shall Space Flight Center. Further infor-mation is contained in a TSP (see page 1).MFS-32081-1


Physical Sciences

A recently proposed boundary condi-tion for atomistic computational model-ing of semiconductor nanostructures(particularly, quantum dots) is an im-proved alternative to two prior suchboundary conditions. As explainedbelow, this boundary condition helps toreduce the amount of computationwhile maintaining accuracy.

The electronic properties of semi-conductor nanostructures (hereaftercalled “nanodevices”) are already uti-lized in sensors, lasers, memory cir-cuits, and electro-optical and optoelec-tronic devices. The electronicproperties of a nanodevice are sensitiveto numerous parameters, includingthose pertaining to sizes, shapes, alloycompositions, and interfaces betweendifferent materials. Atomistic computa-tional simulation of a nanodevice canhelp in the selection of optimal param-eters in the huge design space inhab-ited by the parameters. However, untilnow, the computational burden posedby the large numbers of atoms in a nan-odevice has made it necessary to limitcomputational modeling to a semi-clas-sical, continuum approximation. Thepurpose served by the present bound-ary condition (and by the two priorboundary conditions with which thepresent boundary condition is com-pared) is to enable truncation of thesimulation domain at an artificialboundary surface so that the domaincan be made small enough that atom-istic computational simulation be-comes practical.

The truncation problem can be sum-marized as follows: Whereas, as its namesuggests, a nanodevice can have charac-teristic dimensions of the order ofnanometers, it is typically embeddedwithin a larger semiconductor structurehaving characteristic dimensions of theorder of micrometers. Therefore, in theabsence of a means of truncation, thesimulation domain must typically en-compass all of the atoms containedwithin a micrometer-sized region. Thekey to truncation lies in recognition thata smaller electronically active region is

defined by localization of the electrondensity in and near a potential well es-tablished by the energy-band offset be-tween two adjacent semiconductor ma-terials. However, without a properboundary condition, simply drawing anartificial boundary surface around theelectronically active region results inmany spurious quantum states associ-ated with dangling interatomic bonds atthe boundary surface.

The present boundary condition ef-fectively eliminates the spurious surfacequantum states by artificially shiftingtheir energy levels well above the energyband of interest, as though the danglingbonds were passivated by high-energymolecules. The size of the dangling-bond energy shift is not critical, as long

as it suffices to remove all spurious quan-tum states from the semiconductor bandgap in the electronically active region ofinterest. For example, in the case of anInAs self-assembled quantum dot em-bedded in GaAs, a shift of 5 eV is suffi-cient to remove the spurious states andmake electron and hole energies con-verge to within a few meV (see figure).

The present boundary condition is arefined version of one of the two priorboundary conditions in which the or-bital energies of surface atoms areraised. Whereas the prior boundary con-dition does not differentiate amongsuch details of the surface atoms as thenumbers and directions of their dan-gling bonds, the present boundary con-dition does. The present boundary con-

Boundary Condition for Modeling Semiconductor NanostructuresSimulation domains are truncated without introducing spurious surface quantum states.NASA’s Jet Propulsion Laboratory, Pasadena, California

Eigenvalues of Quantum States in the band gap of a simulated InAs-in-GaAs quantum dot are alreadyconverged within a few meV at a dangling-bond shift of 5 eV, and converge further as the dangling-bond shift is increased to 20 eV.

0

–1

–2

–3

–4

–5

–6

–7 5 10 15 20

Eig

enva

lue

Dif

fere

nce

,meV

Dangling-Bond Energy Shift, eV

NPO-41155 ABPI

2-28-05 es

Hole Ground

State Hole

Excited State

Electron Ground

State

Electron Excited State


dition is more physically realistic be-cause, unlike in the prior boundary con-dition, the connected-bond energy ofthe surface atoms is kept unchangedand, hence, there is no extra energypenalty for electrons to occupy the con-

nected bonds of surface atoms. Theother prior boundary condition is a peri-odic one and, hence, not well suited tomodeling a nanodevice that has an irreg-ular shape or is subjected to a non-peri-odic externally applied potential.

This work was done by Seungwon Lee,Fabiano Oyafuso, Paul von Allmen, andGerhard Klimeck of Caltech for NASA’s JetPropulsion Laboratory. Further informa-tion is contained in a TSP (see page 1).NPO-41155

Miniature Distillation Column for Producing LOX From Air This column is only about a hundredth as high as an industrial one. John H. Glenn Research Center, Cleveland, Ohio

The figure shows components of adistillation column intended for use aspart of a system that produces high-pu-rity liquid oxygen (LOX) from air bydistillation. (The column could be eas-ily modified to produce high-purity liq-uid nitrogen.) Whereas typical indus-trial distillation columns for producinghigh-purity liquid oxygen and/or nitro-gen are hundreds of feet tall, this distil-lation column is less than 3 ft (less thanabout 0.9 m) tall. This column was de-veloped to trickle-charge a LOX-basedemergency oxygen system (EOS) for alarge commercial aircraft.

A description of the industrial produc-tion of liquid oxygen and liquid nitrogenby distillation is prerequisite to a mean-ingful description of the present minia-turized distillation column. Typically,such industrial production takes place ina chemical processing plant in whichlarge quantities of high-pressure air areexpanded in a turboexpander to (1) re-cover a portion of the electrical power re-quired to compress the air and (2) par-tially liquefy the air. The resultingtwo-phase flow of air is sent to the middleof a distillation column. The liquid phaseis oxygen-rich, and its oxygen purity in-creases as it flows down the column. Thevapor phase is nitrogen-rich and its nitro-gen purity increases as it flows up the col-umn. A heater or heat exchanger, com-monly denoted a reboiler, is at thebottom of the column. The reboiler is sonamed because its role is to reboil someof the liquid oxygen collected at the bot-tom of the column to provide a flow ofoxygen-rich vapor. As the oxygen-richvapor flows up the column, it absorbs thenitrogen in the down-flowing liquid bymass transfer. Once the vapor leaves thelower portion of the column, it interactswith down-flowing nitrogen liquid thathas been condensed in a heat exchanger,commonly denoted a condenser, at thetop of the column. Liquid oxygen andliquid nitrogen products are obtained by

draining some of the purified product atthe bottom and top of the column, re-spectively.

Because distillation is a mass-transferprocess, the purity of the product(s) canbe increased by increasing the effective-ness of the mass-transfer process (increas-ing the mass-transfer coefficient) and/orby increasing the available surface areafor mass transfer through increased col-umn height. The diameter of a distilla-tion column is fixed by pressure-drop andmass-flow requirements. The approachtaken in designing the present distillationcolumn to be short yet capable of yield-ing a product of acceptably high puritywas to pay careful attention to design de-tails that affect mass-transfer processes.

The key components in this column arethe structured packing and the distribu-tor. The structured packing is highly com-pact. Each section of packing is about 1in. (about 2.5 cm) in diameter and 3 in.(about 7.6 cm) long. The column con-tains a total of seven sections of packing,so the total length of packing in the col-umn is 21 in. (about 53 cm). The packingpromotes transfer of mass between theup-flowing vapor and the down-flowingliquid. The liquid distributor, as its namesuggests, helps to distribute the liquid asnearly evenly as possible throughout thecross section of the column so as to utilizethe packing to the fullest extent possibleand thereby maximize the mass-transfereffectiveness of the column.

In operation, saturated air at a pres-sure of 70 psia (absolute pressure of0.48 MPa) enters the reboiler and par-tially condenses. The air is then fullycondensed by an external refrigerationsource, such as a small cryocooler. Theair then goes through a pressure dropof about 50 psi (about 0.34 MPa) in athrottling valve and thereby becomespartially vaporized. This pressure dropsets the column pressure at about 20psia (about 0.14 MPa). This columnpressure is required to obtain a signifi-

cant temperature difference in the re-boiler. The two-phase flow then enters aseparator, where the vapor is vented,and the liquid is sent to the distributor.Once operation has reached a steadystate, mass transfer between the down-flowing liquid and the up-flowing vapor

COMPONENTS OF DISTILLATION COLUMN

MAGNIFIED VIEW OF ONE SECTION OF STRUCTURED PACKING

LiquidDistributor

Seven Sectionsof Structured

Packing

Liquid-Level Probe

Reboiler

Column HousingColumn Support

LEW-17593-1ABPI

08-31-04 LE

The Components of the Distillation Column aredesigned to maximize mass transfer in a small space.


results in the collection of 99-percent-pure LOX in the reboiler. The nitrogen-rich vapor is vented as waste at the topof the column. The structured packingenables the column operation to be in-sensitive to tilt angles of up to 20°, withrespect to the local gravity vector. We

are currently working to further minia-turize the distillation technology to pro-vide a portable, lightweight, and low-power source of high-purity nitrogenand oxygen for other applications.

This work was done by Jay C. Rozzi ofCreare, Inc., for Glenn Research Center.

Inquiries concerning rights for the com-mercial use of this invention should be ad-dressed to NASA Glenn Research Center,Commercial Technology Office, Attn: SteveFedor, Mail Stop 4–8, 21000 BrookparkRoad, Cleveland, Ohio 44135. Refer toLEW-17593-1.

Even Illumination From Fiber-Optic-Coupled Laser DiodesEmerging light beams would be shaped by diffractive fiber-optic tips.Marshall Space Flight Center, Alabama

A method of equipping fiber-optic-coupled laser diodes to evenly illumi-nate specified fields of view has beenproposed. The essence of the method isto shape the tips of the optical fibersinto suitably designed diffractive opticalelements. One of the main benefits af-forded by the method would be morenearly complete utilization of the avail-able light.

As shown in Figure 1, the light beamemerging from the flat tip of an opticalfiber coupled to a laser diode has aGaussian distribution of intensity across acircular cross section, whereas what istypically desired is to concentrate thelight into a beam characterized by a “top-hat” distribution (even illumination in aspecified field of view, zero illuminationoutside the field of view). In order to ob-tain an acceptably close approximationof even illumination in the field of view,the Gaussian beam must be significantlywider, so that much or most of the lightis wasted outside the field of view. A con-ventional lens can be used to partiallyshape the beam, but the beam does notlose its basic Gaussian character; this istrue whether the lens is placed at a focaldistance from the tip, in contact with thetip, or formed onto the tip surface as anintegral part of the optical fiber.

Diffractive optics is a relatively newfield of optics in which laser beams areshaped by use of diffraction instead of re-fraction. There exist ways to produce dif-fractive lens elements that shape laserbeams into desired arbitrary cross sec-tions (for example, the arrow shapes ofthe beams generated by many laser point-ers). In a fiber-optic-coupled laser diodeaccording to the proposal, the opticalfiber would be tipped with a diffractivesurface such that the diffraction patternimposed on light leaving the fiber would,at a desired distance from the tip, con-centrate the beam at nearly even intensityinto a cross section of specified shape.

Usually, the desired illuminated areawould be rectangular or circular, but inprinciple, the diffractive surface could bedesigned to shape the beam to almost anyspecified cross section. In one version ofthe proposal, the diffractive shape wouldbe etched directly onto the initially flattip surface of the fiber. In another ver-sion, the diffractive surface would bemolded onto a transparent piece of plas-tic that would be bonded to the tip, themold having been previously etched orotherwise formed to the diffractive shape.

A diffractive fiber-tip surface thatwould function in this way has not yetbeen designed. However, it has been es-timated, for example, that such a pat-tern on the tip of an optical fiber of 110-µm diameter would consist of about 300prisms of various heights resemblingbuildings on 5-µm-square city blocks(see Figure 2), fabricated by etching thesquare areas to different depths from aninitial flat tip surface. A small develop-mental problem is posed by the diffi-culty of etching such a pattern.

As in the case of any diffractiveoptic, some light would pass throughundiffracted; hence, the output lightpattern that would be mostly the de-sired pattern with a slight superim-posed Gaussian pattern.

This work was done by Richard T. Howard ofMarshall Space Flight Center. Further infor-mation is contained in a TSP (see page 1).

This invention is owned by NASA, and apatent application has been filed. For furtherinformation, contact Sammy Nabors, MSFCCommercialization Assistance Lead, [email protected]. Refer to MFS-31843-1.

–30 0

0.2

0.4

0.6

0.8

1.0

1.2

–20 –10 0

Half Cone Angle, Degrees

Ou

tpu

tIn

ten

sity

10 20 30

Desired "Top Hat"

Output

Unmodified Gaussian Output

MFS31843-1 Fig 1ABPI

5-24-05 bs

Figure 1. The Distribution of Laser Light emerg-ing from an optical fiber is typically Gaussian,whereas often a “top hat” distribution is desired.

MFS31843-1 Fig 2 ABPI

5-24-05 bs

Figure 2. This Diffractive Optic was formed in plastic. Optical performance is affected by the widthsand depths of steps and the sharpness of edges. Although a rectangular floor plan is shown in this ex-ample, the floor plan for application to the tip of a round optical fiber would be circular, even if theoptic were to be used to illuminate a rectangular area.


Optically driven deformable mirrors mayeventually supplant electrically driven de-formable mirrors in some adaptive-opticsand active-optics applications. Traditionally,the mirror facets in electrically driven de-formable mirrors are actuated, variously, bymeans of piezoelectric, electrostrictive, mi-croelectromechanical, liquid-crystal, orthermal devices. At least one such devicemust be dedicated to each facet, and theremust be at least one wire carrying a controlor drive signal to the device. If a deformablemirror comprises many (e.g., thousands) offacets, then wiring becomes a major prob-lem for design, and the problem is com-pounded in cases of piezoelectric or otheractuators for which high drive voltages arerequired. In contrast, in optically drivenmirrors, the wiring problem is eliminated.

The basic principle of actuation of anoptically driven deformable mirror is touse a laser beam to actuate a material. Forexample, a laser beam can be used toheat a material to make the material ther-mally expand to displace a mirror facet.In an experiment to demonstrate thisprinciple, the actuator was a Golay cell(see Figure 1) having a diameter of ≈6mm and a length of ≈10 mm. The beamfrom a laser diode was aimed at an ab-sorber in the cell, thereby heating the gasin the cell. A mirror mounted on a 12.5-µm-thick polyethylene terephthalate di-aphragm at one end of the cell became

displaced as the gas expanded against thediaphragm. In one representative pair ofexperiments at a laser beam power of0.23 W, the beam was mechanicallychopped at frequencies of 1 and 5 Hz.The mirror exhibited corresponding os-cillating displacements having ampli-tudes of 373 and 83 µm, respectively.

Figure 2 depicts a simple experimentaldeformable mirror comprising a 5×5square array of Golay cells with square mir-ror facets mounted on their membranes. Atypical practical deformable mirror wouldlikely include a much larger array (e.g.,100×100). In the contemplated use of suchan array, two computer-controlled single-axis mirrors would be used to raster-scan alaser beam across the array, and the rasterscan would be synchronized with an ampli-tude modulation to control the amount of

heat delivered to each cell and thereby tocontrol the displacement of each facet.

This work was done by Hamid Hemmati andWilliam Farr of Caltech for NASA’s Jet Propul-sion Laboratory. Further information is con-tained in a TSP (see page 1).

In accordance with Public Law 96-517, thecontractor has elected to retain title to this inven-tion. Inquiries concerning rights for its commer-cial use should be addressed to:

Innovative Technology Assets ManagementJPLMail Stop 202-2334800 Oak Grove DrivePasadena, CA 91109-8099(818) 354-2240E-mail: [email protected] to NPO-42724, volume and number of

this NASA Tech Briefs issue, and the pagenumber.

Optically Driven Deformable MirrorsThere is no wiring on the back sides of these mirrors.NASA’s Jet Propulsion Laboratory, Pasadena, California

Golay Cell

Window

Absorber

Chopper

Membrane

Small Mirror Mounted on Membrane

Laser Diode

NPO-42724 Fig 1ABPI

11-18-05 bs

Figure 1. Light From a Laser Diode impinges on an absorber in a Golay cell, heating the gas in the cell.A mirror on the diaphragm is displaced by the resulting expansion of the gas.


Information Sciences

Algorithm for Automated Detection of Edges of CloudsThe algorithm has been shown to be reliable and robust.John F. Kennedy Space Center, Florida

An algorithm processes cloud-physicsdata gathered in situ by an aircraft,along with reflectivity data gathered byground-based radar, to determinewhether the aircraft is inside or outsidea cloud at a given time. A cloud edge isdeemed to be detected when thein/out state changes, subject to a hys-teresis constraint. Such determinationsare important in continuing researchon relationships among lightning, elec-tric charges in clouds, and decay ofelectric fields with distance from cloudedges.

More specifically, the algorithm con-sists of an in-cloud detection componentand a boundary detection component.The in-cloud detection component relieson the cloud-physics and weather-radardata to make a tentative determination ofthe in/out state. The boundary detectioncomponent examines the output of thein-cloud detection component and ap-plies a hysteresis test, which helps preventfalse boundary detections that would oth-erwise be triggered by momentary datafluctuations associated with isolated tran-sient cloud puffs or data dropouts.

The algorithm was tested by applyingit to a large set of data and comparingthe results of the algorithm with resultsobtained through detailed manual ex-amination of the data. The algorithmwas found to be highly reliable and in-sensitive to transient instrumentationnoise or data gaps, and it enabled fullautomation of detection of cloud edges.

This work was done by Jennifer G. Wardand Francis J. Merceret of Kennedy SpaceCenter. Further information is contained ina TSP (see page 1)..KSC-12574

Exploiting Quantum Resonance to Solve CombinatorialProblemsNASA’s Jet Propulsion Laboratory, Pasadena, California

Quantum resonance would be ex-ploited in a proposed quantum-comput-ing approach to the solution of combi-natorial optimization problems. Inquantum computing in general, onetakes advantage of the fact that an algo-rithm cannot be decoupled from thephysical effects available to implementit. Prior approaches to quantum com-puting have involved exploitation ofonly a subset of known quantum physi-cal effects, notably including parallelism

and entanglement, but not includingresonance. In the proposed approach,one would utilize the combinatorialproperties of tensor-product decompos-ability of unitary evolution of many-par-ticle quantum systems for physically sim-ulating solutions to NP-completeproblems (a class of problems that areintractable with respect to classicalmethods of computation). In this ap-proach, reinforcement and selection ofa desired solution would be executed by

means of quantum resonance. Classes ofNP-complete problems that are impor-tant in practice and could be solved bythe proposed approach include plan-ning, scheduling, search, and optimaldesign.

This work was done by Michail Zak andAmir Fijany of Caltech for NASA’s JetPropulsion Laboratory. Further informa-tion is contained in a TSP (see page 1).NPO-41902

Hybrid Terrain DatabaseAerial photographs are draped onto digital elevation maps.Langley Research Center, Hampton, Virginia

A prototype hybrid terrain database isbeing developed in conjunction withother databases and with hardware andsoftware that constitute subsystems ofaerospace cockpit display systems(known in the art as synthetic vision sys-tems) that generate images to increase

pilots’ situation awareness and eliminatepoor visibility as a cause of aviation acci-dents. The basic idea is to provide a clearview of the world around an aircraft bydisplaying computer-generated imageryderived from an onboard database of ter-rain, obstacle, and airport information.

The hybrid terrain database, whichcould constitute all or part of such anonboard database, can be characterizedas an accurate model of terrain and ob-stacles of interest to a pilot. The hybridterrain database contains (1) imageryderived from conventional aerial photo-


graphs of a terrain area of interest, (2) adigital elevation map of the terrain areaof interest, and, optionally, (3) informa-tion on the area of interest from otherdatabases.

The hybrid terrain database is used togenerate synthetic terrain imagery (see fig-ure) in a hybrid textured format, which isbest described in terms of the process bywhich the synthetic imagery is generated:The hybrid texture is created from mono-chromatic aerial photographs of the af-fected terrain area merged with color-coded elevation-based terrain imagery (seefigure). The process involves coloring themonochromatic aerial photographs of theaffected area on the basis of altitude anddraping the resulting elevation-coloredphotographs onto the digital elevationmodel of the terrain.

This work was done by Trey Arthur of Lang-ley Research Center. Further information iscontained in a TSP (see page 1).LAR-16898-1

This Hybrid Textured Terrain image was generated from a hybrid terrain database of Reno-Tahoe In-ternational Airport and its environs.

LAR-16898

ABPI

02-01-06 LE


Books & Reports

On Release of Microbe-Laden Particles From MarsLanders

A paper presents a study in whichrates of release of small particles fromMars lander spacecraft into the Martianatmosphere were estimated from firstprinciples. Because such particles canconsist of, or be laden with, terrestrialmicrobes, the study was undertaken tounderstand their potential for biologicalcontamination of Mars. The study in-cluded taking account of forces and en-ergies involved in adhesion of particlesand of three mechanisms of dislodge-ment of particles from the surface of aMars lander: wind shear, wind-driven im-pingement of suspended dust, and im-pingement of wind-driven local saltatingsand particles. Wind shear was deter-mined to be effective in dislodging onlyparticles larger than about 10 micronsand would probably be of limited inter-est because such large particles could beremoved by pre-flight cleaning of thespacecraft, and their number on thelaunched spacecraft would thus be rela-tively small. Dislodge-ment by wind-driven dust was found to be character-ized by an adhesion half-life of the orderof 10,000 years — judged to be too longto be of concern. Dislodgement by saltat-ing sand particles, including skirts ofdust devils, was found to be of potentialimportance, depending on the sizes ofthe spacecraft-attached particles andcharacteristics of both Mars sand-parti-cle and spacecraft surfaces.

This work was done by Josette Bellan andKenneth Harstad of Caltech for NASA’s JetPropulsion Laboratory. Further informa-tion is contained in a TSP (see page 1).NPO-42687

A Concept for Run-TimeSupport of the Chapel Language

A document presents a concept forrun-time implementation of other con-cepts embodied in the Chapel program-ming language. (Now undergoing devel-opment, Chapel is intended to become astandard language for parallel comput-ing that would surpass older such lan-guages in both computational perform-ance in the efficiency with whichpre-existing code can be reused and new

code written.) The aforementionedother concepts are those of distribu-tions, domains, allocations, and access,as defined in a separate document called“A Semantic Framework for Domainsand Distributions in Chapel” and linkedto a language specification defined inanother separate document called“Chapel Specification 0.3.” The conceptpresented in the instant report is recog-nition that a data domain that was in-vented for Chapel offers a novel ap-proach to distributing and processingdata in a massively parallel environment.The concept is offered as a starting pointfor development of working descriptionsof functions and data structures thatwould be necessary to implement inter-faces to a compiler for transforming theaforementioned other concepts fromtheir representations in Chapel sourcecode to their run-time implementations.

This work was done by Mark James of Cal-tech for NASA’s Jet Propulsion Labora-tory. Further information is contained in aTSP (see page 1).

The software used in this innovation isavailable for commercial licensing. Pleasecontact Karina Edmonds of the CaliforniaInstitute of Technology at (626) 395-2322.Refer to NPO-42496.

Thermoelectric Inhomogeneities in(Ag1–ySbTe2)x(PbTe)1–x

A document presents a study of whymaterials of composition (Ag1–ySbTe2)0.05

(PbTe)0.95 [0≤y≤1] were previously re-ported to have values of the thermoelec-tric figure of merit [ZT (where Z ≡α2/ρκ, α is the Seebeck coefficient, ρ iselectrical resistivity, κ is thermal conduc-tivity, and T is absolute temperature)]ranging from <1 to >2. In the study, sam-ples of (AgSbTe2)0.05(PbTe)0.95,(Ag0.67SbTe2)0.05 (PbTe)0.95, and(Ag0.55SbTe2)0.05(PbTe)0.95 were pre-pared by melting followed, variously, byslow or rapid cooling. Analyses of thesesamples by x-ray diffraction, electron mi-croscopy, and scanning-microprobemeasurements of the Seebeck coeffi-cient led to the conclusion that thesematerials have a multiphase characteron a scale of the order of millimeters,even though they appear homogeneousin x-ray diffraction and electron mi-croscopy. The Seebeck measurements

showed significant variations, includingboth n-type and p-type behavior in thesame sample. These variations werefound to be consistent with observedvariations of ZT. The rapidly quenchedsamples were found to be less inhomo-geneous than were the furnace-cooledones; hence, rapid quenching was sug-gested as a basis of research on synthesiz-ing more nearly uniform high-ZT sam-ples.

This work was done by G. Jeffrey Snyder,Nancy Chen, Franck Gascoin, EckhardMueller, Gabriele Karpinski, and ChristianStiewe of Caltech for NASA’s Jet PropulsionLaboratory. Further information is con-tained in a TSP (see page 1).NPO-42657

Spacecraft Escape CapsuleA report discusses the Gumdrop cap-

sule — a conceptual spacecraft thatwould enable the crew to escape safely inthe event of a major equipment failureat any time from launch through atmos-pheric re-entry. The scaleable Gumdropcapsule would comprise a commandmodule (CM), a service module (SM),and a crew escape system (CES). TheCM would contain a pressurized crewenvironment that would include avionic,life-support, thermal control, propulsiveattitude control, and recovery systems.The SM would provide the primarypropulsion and would also supply elec-trical power, life-support resources, andactive thermal control to the CM. TheCES would include a solid rocket motor,embedded within the SM, for pushingthe CM away from the SM in the event ofa critical thermal-protection-system fail-ure or loss of control. The CM and SMwould normally remain integrated witheach other from launch through recov-ery, but could be separated using theCES, if necessary, to enable the safe re-covery of the crew in the CM. The crewescape motor could be used, alterna-tively, as a redundant means of de-orbitpropulsion for the CM in the event of amajor system failure in the SM.

This work was done by Edward A. Robert-son, Dingell W. Charles, Ann L. Bufkin,Liana M. Rodriggs, Wayne Peterson, PeterCuthbert, David E. Lee, and Carlos Westhelleof Johnson Space Center. Further informa-tion is contained in a TSP (see page 1).MSC-23840

National Aeronautics andSpace Administration

Technology Focus Electronics/Computers · 11 Photodetector Arrays for Multicolor Visible/Infrared Imaging 12 Semiconductor Bolometers Give Background- ... 22 Solar-Powered Cooler

Documents