technology focus electronics/computers - ntrs.nasa.gov · vectorized rebinning algorithm for fast...

Technology Focus

Electronics/Computers

Software

Materials

Mechanics/Machinery

Manufacturing

Bio-Medical

Physical Sciences

Information Sciences

Books and Reports

3-13 March 2013

https://ntrs.nasa.gov/search.jsp?R=20130011219 2019-05-05T22:40:44+00:00Z

NASA Tech Briefs, March 2013 1

INTRODUCTIONTech Briefs are short announcements of innovations originating from research and developmentactivities of the National Aeronautics and Space Administration. They emphasize information con-sidered likely to be transferable across industrial, regional, or disciplinary lines and are issued toencourage commercial application.

Additional Information on NASA Tech Briefs and TSPsAdditional information announced herein may be obtained from the NASA Technical Reports Server:http://ntrs.nasa.gov.

Please reference the control numbers appearing at the end of each Tech Brief. Infor mation on NASA’s Innovative Partnerships Program (IPP), its documents, and services is available on the World Wide Webat http://www.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 CenterDavid Morse(650) [email protected]

Dryden Flight Research CenterRon Young(661) [email protected]

Glenn Research CenterKimberly A. Dalgleish-Miller(216) [email protected]

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

Jet Propulsion LaboratoryDan Broderick(818) [email protected]

Johnson Space CenterJohn E. James(281) [email protected]

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

Langley Research CenterMichelle Ferebee(757) [email protected]

Marshall Space Flight CenterTerry L. Taylor(256) [email protected]

Stennis Space CenterRamona Travis(228) [email protected]

NASA Headquarters

Daniel Lockney, Technology Transfer Program Executive(202) [email protected]

Small Business Innovation Research(SBIR) & Small Business TechnologyTransfer (STTR) ProgramsRich Leshner, Program Executive(202) [email protected]

NASA Field Centers and Program Offices

5 Technology Focus: Data Acquisition

5 Remote Data Access with IDL5 Data Compression Algorithm Architecture for Large

Depth-of-Field Particle Image Velocimeters5 Vectorized Rebinning Algorithm for Fast Data

Down-Sampling6 Display Provides Pilots with Real-Time

Sonic-Boom Information 6 Onboard Algorithms for Data Prioritization and

Summarization of Aerial Imagery7 Monitoring and Acquisition Real-time System (MARS)

9 Electronics/Computers9 Analog Signal Correlating Using an

Analog-Based Signal Conditioning Front End 9 Micro-Textured Black Silicon Wick for Silicon

Heat Pipe Array10 Robust Multivariable Optimization and Performance

Simulation for ASIC Design

13 Manufacturing & Prototyping13 Castable Amorphous Metal Mirrors

and Mirror Assemblies

15 Mechanics/Machinery15 Sandwich Core Heat-Pipe Radiator for Power

and Propulsion Systems16 Apparatus for Pumping a Fluid17 Cobra Fiber-Optic Positioner Upgrade

19 Materials & Coatings19 Improved Wide Operating Temperature

Range of Li-Ion Cells 19 Non-Toxic, Non-Flammable, –80 ºC Phase

Change Materials20 Soft-Bake Purification of SWCNTs Produced by

Pulsed Laser Vaporization

23 Bio-Medical23 Improved Cell Culture Method for Growing

Contracting Skeletal Muscle Models24 Hand-Based Biometric Analysis24 The Next Generation of Cold Immersion Dry Suit

Design Evolution for Hypothermia Prevention

27 Software27 Integrated Lunar Information Architecture for

Decision Support Version 3.0 (ILIADS 3.0) 27 Relay Forward-Link File Management Services

(MaROS Phase 2)27 Two Mechanisms to Avoid Control Conflicts

Resulting from Uncoordinated Intent28 XTCE GOVSAT Tool Suite 1.0

29 Information Technology29 Determining Temperature Differential to

Prevent Hardware Cross-Contamination in a Vacuum Chamber

29 SequenceL: Automated Parallel Algorithms Derivedfrom CSP-NT Computational Laws

30 Remote Data Exploration with the Interactive Data Language (IDL)

30 Mixture-Tuned, Clutter Matched Filter for RemoteDetection of Subpixel Spectral Signals

31 Physical Sciences31 Partitioned-Interval Quantum Optical

Communications Receiver32 Practical UAV Optical Sensor Bench With Minimal

Adjustability

3-13 March 2013


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.


Technology Focus: Data Acquisition

Vectorized Rebinning Algorithm for Fast Data Down-SamplingApplications include image processing, filter design, and anti-aliasing techniques.Goddard Space Flight Center, Greenbelt, Maryland

A vectorized rebinning (down-sam-pling) algorithm, applicable to N-di-mensional data sets, has been developedthat offers a significant reduction incomputer run time when compared toconventional rebinning algorithms. Forclarity, a two-dimensional version of thealgorithm is discussed to illustrate somespecific details of the algorithm content,and using the language of image pro-

cessing, 2D data will be referred to as“images,” and each value in an image asa “pixel.” The new approach is fully vec-torized, i.e., the down-sampling proce-dure is done as a single step over allimage rows, and then as a single stepover all image columns.Data rebinning (or down-sampling) is

a procedure that uses a discretely sam-pled N-dimensional data set to create a

representation of the same data, but withfewer discrete samples. Such data down-sampling is fundamental to digital signalprocessing, e.g., for data compression ap-plications. Additional applications in-clude image processing, filter design, andanti-aliasing techniques. Data rebinningis a computationally intensive procedureand thus the goal in this technology de-velopment is a more efficient algorithm

A large depth-of-field particle imagevelocimeter (PIV) is designed to char-acterize dynamic dust environmentson planetary surfaces. This instrumentdetects lofted dust particles, andsenses the number of particles per unitvolume, measuring their sizes, veloci-ties (both speed and direction), andshape factors when the particles arelarge. To measure these particle char-acteristics in-flight, the instrumentgathers two-dimensional image data ata high frame rate, typically >4,000 Hz,generating large amounts of data forevery second of operation, approxi-mately 6 GB/s.

To characterize a planetary dust envi-ronment that is dynamic, the instru-ment would have to operate for at leastseveral minutes during an observationperiod, easily producing more than aterabyte of data per observation. Givencurrent technology, this amount ofdata would be very difficult to store on-board a spacecraft, and downlink toEarth. Since 2007, innovators havebeen developing an autonomous imageanalysis algorithm architecture for thePIV instrument to greatly reduce theamount of data that it has to store anddownlink. The algorithm analyzes PIVimages and automatically reduces the

image information down to only theparticle measurement data that is of in-terest, reducing the amount of datathat is handled by more than 103. Thestate of development for this innova-tion is now fairly mature, with a func-tional algorithm architecture, alongwith several key pieces of algorithmlogic, that has been proven throughfield test data acquired with a proof-of-concept PIV instrument.This work was done by Brent Bos, Nargess

Memarsadeghi, Semion Kizhner, and ScottAntonille of Goddard Space Flight Center.Further information is contained in a TSP(see page 1). GSC-15960-1

Data Compression Algorithm Architecture for Large Depth-of-Field Particle Image VelocimetersGoddard Space Flight Center, Greenbelt, Maryland

Remote Data Access with IDLGoddard Space Flight Center, Greenbelt, Maryland

A tool based on IDL (Interactive DataLanguage) and DAP (Data Access Proto-col) has been developed for user-friendly remote data access. A difficultyfor many NASA re searchers using IDL isthat often the data to analyze are locatedremotely and are too large to transferfor local analysis. Researchers have de-veloped a protocol for accessing remotedata, DAP, which is used for both SOHOand STEREO data sets. Server-side side

analysis via IDL routine is availablethrough DAP.The tools allow normal DAP users to

run IDL scripts on their data remotelyvia DAP. This powerful, user-friendly in-terface to DAP for IDL improved OPeN-DAP bindings that fixed bugs in existingfunctionality, created a GUI client to ex-plore data sets served with DAP, devel-oped a pure IDL DAP implementationthat provided complete DAP capabilities

along with a simple installation, im-proved network capabilities for GDL(the open source IDL alternative) andolder versions of IDL, and modified theOPeNDAP Hyrax DAP server to processdata on the server-side via a syntax in theDAP request.This work was done by Michael Galloy of

Tech-X Corporation for Goddard Space FlightCenter. Further information is contained in aTSP (see page 1). GSC-16253-1

6 NASA Tech Briefs, March 2013

Many current and future NASA mis-sions are capable of collecting enormousamounts of data, of which only a smallportion can be transmitted to Earth.Communications are limited due to dis-tance, visibility constraints, and compet-ing mission downlinks. Long missionsand high-resolution, multispectral imag-

ing devices easily produce data exceed-ing the available bandwidth. To addressthis situation, computationally efficientalgorithms were developed for analyzingscience imagery onboard the spacecraft.These algorithms autonomously clusterthe data into classes of similar imagery,enabling selective downlink of represen-

tatives of each class, and a map classify-ing the terrain imaged rather than thefull dataset, reducing the volume of thedownlinked data. A range of approacheswas examined, including k-means clus-tering using image features based oncolor, texture, temporal, and spatialarrangement.

Onboard Algorithms for Data Prioritization and Summarizationof Aerial ImageryClustering/machine learning methods are used to structure data for prioritization, mapping,and downlinking.NASA’s Jet Propulsion Laboratory, Pasadena, California

with reduced run times, as compared toexisting rebinning approaches. This ap-proach is able to take advantage of vector-ized instructions such as Single Instruc-tion Multiple Data (SIMD), to performthe rebinning operation.The algorithm completely vectorizes

the data rebinning operation, in thesense that a “single” arithmetic opera-tion is applied simultaneously to multi-ply distinct data sets and is executed withthe approximate run time of that opera-tion applied to a single data set. Forlower-level computer languages, such asC or assembly, vectorized operations can

be implemented using central process-ing unit (CPU) single-instruction, multi-ple-data (SIMD) capabilities, such asstreaming SIMD Extensions 3 (SSE3) onx86 computer architecture or AltiVec onPowerPC processors. Thus, although thealgorithm has been implemented usingMATLAB, it is not fundamentally tied toMATLAB, and can be implementedusing other programming languages.The vectorized data rebinning (down-

sampling) procedure offers a reducedrun time when compared with standardrebinning algorithms. In general, algo-rithms are often optimized by trading

decreased run time for increased mem-ory, where the latter is needed for stor-ing additional code, pre-computed re-sults, or other ancillary data. However,the vectorized rebinning approach doesnot have increased memory require-ments compared with conventional ap-proaches. The underlying fundamentaladvantage to this technology is the uti-lization of vectorized instructions for therebinning operation.This work was done by Bruce Dean, David

Aronstein, and Jeffrey Smith of Goddard SpaceFlight Center. Further information is con-tained in a TSP (see page 1). GSC-15949-1

Supersonic aircraft generate shockwaves that move outward and extend tothe ground. As a cone of pressurized airspreads across the landscape along theflight path, it creates a continuous sonicboom along the flight track. Several fac-tors can influence sonic booms: weight,size, and shape of the aircraft; its alti-tude and flight path; and weather andatmospheric conditions. This technol-ogy allows pilots to control the impactof sonic booms. A software system displays the location

and intensity of shock waves caused bysupersonic aircraft. This technology canbe integrated into cockpits or flight con-trol rooms to help pilots minimize sonic-boom impact in populated areas. Thesystem processes vehicle and flight pa-rameters as well as data regarding cur-

rent atmospheric conditions. The dis-play provides real-time information re-garding sonic boom location and inten-sity, enabling pilots to make thenecessary flight adjustments to controlthe timing and location of sonic booms.This technology can be used on current-generation supersonic aircraft, whichgenerate loud sonic booms, as well as fu-ture-generation, low-boom aircraft, an-ticipated to be quiet enough for popu-lated areas. When fully deployed in real time, the

display will leverage existing tools devel-oped and enhanced by the U.S. AirForce and NASA to predict sonic boomparameters. The prediction data will beintegrated with a real-time, local-area,moving-map display that is capable ofdisplaying the aircraft’s current sonic

boom footprint at all times. The pilotwill be able to choose from a menu ofpre-programmed maneuvers such as ac-celerations, turns, or pushovers, and thepredicted sonic boom footprint for thatmaneuver appears on the map. Afterfully developed and implemented, thiswill allow the pilot to select or modify pa-rameters to either avoid generating asonic boom or to place the sonic boomin a specific location. The system mayalso provide pilots with guidance on howto execute the chosen maneuver. This technology will enable super-

sonic commercial flight without disturb-ing population centers on the ground. This work was done by Ed Haering of Dry-

den Flight Research Center and Ken Plotkin ofWyle. Further information is contained in aTSP (see page 1). DRC-008-001

Display Provides Pilots with Real-Time Sonic-Boom Information The impact of sonic booms can be controlled over populated areas. Dryden Flight Research Center, Edwards, California


Several unique challenges influenceddesign decisions for automatic imageanalysis. First, onboard processing is lim-ited in spaceflight applications. Avionicscomputers must satisfy strict radiationand energy constraints, and their re-sources are shared between continuousautonomous control and data process-ing. Computational constraints mandatea simple approach to image analysis inwhich statistical properties of the imageserve as proxies for the actual content.A major challenge is the diversity of

surface features an aerobot might en-counter. An aerobot would be in con-stant motion but difficult to control dueto unpredictable atmospheric currents.It would be difficult to schedule imagetargets in advance or to anticipate thefeatures of interest that will appear. This

favors an “unsupervised” approach thatmakes few assumptions about imagecontent but instead discovers interestingand representative samples based on theintrinsic properties of the data. Cluster-ing is one common unsupervised ap-proach; it classifies a dataset into dis-crete categories of items with similarproperties.Image features can be considered to

fall into one of four groups, or themes,based on the properties they describe.These are color, edge, frequency, andtime. The edge and frequency featurescorrelate with image texture, color cap-tures basic color statistics, and time de-scribes the temporal order in which theimages were collected.The main feature of this innovation is

the use of clustering/machine learning

methods to structure data for prioritiza-tion, mapping, and downlink. The effec-tiveness of clustering rests on the qualityof the feature vectors describing each setof data. Features that are redundant, orhave little variance, will reduce the effec-tiveness of clustering. Dimensionality re-duction techniques such as principalcomponent analysis (PCA) can trans-form a high-dimensional feature spaceinto a lower-dimensional space wherethe new, uncorrelated features haveheightened variance. Ideal clusteringscontain compact clusters that are spreadfar apart from one another.This work was done by Steve A. Chien,

David Hayden, David R. Thompson, andRebecca Castano of Caltech for NASA’s JetPropulsion Laboratory. For more information,contact [email protected]. NPO-47534

MARS is a graphical user interface(GUI) written in MATLAB and Java, al-lowing the user to configure and controlthe Scalable Parallel Architecture forReal-Time Acquisition and Analysis(SPARTAA) data acquisition system.SPARTAA not only acquires data, butalso allows for complex algorithms to beapplied to the acquired data in real time.The MARS client allows the user to setup and configure all settings regardingthe data channels attached to the system,as well as have complete control overstarting and stopping data acquisition. Itprovides a unique “Test” programmingenvironment, allowing the user to createtests consisting of a series of alarms, eachof which contains any number of datachannels. Each alarm is configured witha particular algorithm, determining thetype of processing that will be applied oneach data channel and tested against a

defined threshold. Tests can be up-loaded to SPARTAA, thereby teaching ithow to process the data.MARS was developed as a front-end

GUI for setup, control, and plotting ofdata from SPARTAA. The system was de-signed to monitor spectral components inreal time from instrumentation locatedon high-speed rotational hardware (pri-marily high-pressure turbo pumps), and toissue cut commands to a facility if presetlevels were violated. However, the systemis not limited to rotational hardware, andcan be used to monitor any level of fre-quency information from a myriad of in-strumented test hardware. The controlsoftware allows the user to configure thesystem easily to support testing of variousconfigurations with multiple alarms, vot-ing logic, and sensor validation,The uniqueness of MARS is in its ca-

pability to be adaptable easily to many

test configurations. Test hardwaremeasurement limits (i.e. vibration,pressure, temperature, etc.) can be pre-determined, and MARS can be used toset up and support quickly any test con-figuration. Multiple alarms with varioustimings can be configured within min-utes, as opposed to previous softwaremodifications. MARS sends and re-ceives protocols via TCP/IP, which al-lows for quick integration into almostany test environment. The use of MAT-LAB and Java as the programming lan-guages allows for developers to inte-grate the software across multipleoperating platforms.This work was done by Corbin Holland of

Marshall Space Flight Center. For more infor-mation, contact Sammy Nabors, MSFC Com-mercialization Assistance Lead, [email protected]. Refer to MFS-32905-1.

Monitoring and Acquisition Real-time System (MARS)Marshall Space Flight Center, Alabama


Electronics/Computers

Planar, semiconductor heat arrays havebeen previously proposed and developed;however, this design makes use of a novel,microscale black silicon wick structurethat provides increased capillary pumpingpressure of the internal working fluid, re-sulting in increased effective thermal con-ductivity of the device, and also enablesoperation of the device in any orientationwith respect to the gravity vector.In a heat pipe, the efficiency of ther-

mal transfer from the case to the work-ing fluid is directly proportional to thesurface area of the wick in contact with

the fluid. Also, the primary failure mech-anism for heat pipes operating withinthe temperature range of interest is in-adequate capillary pressure for the re-turn of fluid from the condenser to thewick. This is also what makes the opera-tion of heat pipes orientation-sensitive.Thus, the two primary requirements fora good wick design are a large surfacearea and high capillary pressure. Surfacearea can be maximized throughnanomachined surface roughening.Capillary pressure is largely driven bythe working fluid and wick structure.

The proposed nanostructure wickhas characteristic dimensions on theorder of tens of microns, which pro-motes menisci of very small radii. Thisresults in the possibility of enormouspumping potential due to the inverseproportionality with radius. Wetting,which also enhances capillary pump-ing, can be maximized through growthof an oxide layer or material deposi-tion (e.g. TiO2) to create a superhy-drophilic surface.In addition, the wick fabrication tech-

nique produces nanostructure forests that

Micro-Textured Black Silicon Wick for Silicon Heat Pipe ArrayThis technology can be used for microprocessors, power switching circuits, and diode lasers inhigh-power electronics.NASA’s Jet Propulsion Laboratory, Pasadena, California

This innovation is capable of corre-lating two analog signals by using ananalog-based signal conditioning frontend to hard-limit the analog signalsthrough adaptive thresholding into abinary bit stream, then performing thecorrelation using a Hamming “similar-ity” calculator function embedded in aone-bit digital correlator (OBDC). Byconverting the analog signal into a bitstream, the calculation of the correla-tion function is simplified, and lesshardware resources are needed. Thisbinary representation allows the hard-ware to move from a DSP where in-structions are performed serially, intodigital logic where calculations can beperformed in parallel, greatly speedingup calculations. Each of two analog signals (channels

A and B) is converted to a digital bitstream by phase correcting it and com-paring it to an average of itself at asampling clock rate f . The hard-lim-ited conversions of A and B are bitwisecompared to measure the level of sim-ilarity between the two by the OBDC.

This similarity measurement X is equalto the maximum possible Hammingdistance (N bits in disagreement)minus the measured number of bits indisagreement. The OBDC functions are embedded

into a field programmable gate array(FPGA). The OBDC is made up of twoshift registers containing the currentsample values (of length N) from eachof the two input channels (A and B).During each sample clock, a new sam-ple from each A and B input is clockedinto the input linear shift register foreach respective channel; this inputshifts the current values in the linearshift register. The oldest (N + 1 sampleclocks ago) sample is clocked out of theregister. Once the inputs have beenclocked in, the correlation routine canstart. This rising edge of the sampleclock also clears the max correlationvalue, the best correlation index, andthe shift counter registers, initializingthe correlator. When the two registers match exactly,

or are correlated, the X value will equal

N. Once the correlation value has beencalculated, this result is forwarded tocompare with the max correlation valueregister. If the X value is greater thanthe current max correlation value, thenthe max correlation value becomes X,and the shift counter register is latchedand put into the best correlation indexregister, providing the index of the cur-rent best correlation. This index is the number of sample

clock periods difference between thetwo input signals and thus, for sampleclock rate f, indicates the delay betweenthe signals A and B.This work was done by Norman Prokop

and Michael Krasowski of Glenn ResearchCenter. Further information is contained in aTSP (see page 1).Inquiries concerning rights for the com-

mercial use of this invention should be ad-dressed to NASA Glenn Research Center, In-novative Partnerships Office, Attn: StevenFedor, Mail Stop 4–8, 21000 BrookparkRoad, Cleveland, Ohio 44135. Refer toLEW-18902-1.

Analog Signal Correlating Using an Analog-Based SignalConditioning Front End Converting a signal into a bit stream simplifies the correlation function calculation. John H. Glenn Research Center, Cleveland, Ohio

are planar, and can take advantage of 2Dheat spreading over a surface vs. state-of-the-art 1D heat transport associated withheat pipes. The combined result of these

benefits promises to be a two-phase heattransfer device, which is very insensitive toa gravity field. Although liquid pressuredrops may be relatively large depending

on the nanostructure density, the overalldevice dimensions of ≈ 7×7 cm are ex-pected to be well within the overall capil-lary limit. The novel aspects of the cur-rently proposed effort include the use ofthe phenomenon of superhydrophilicityin a heat pipe, and the wick geometry, con-trol of the nanotip height density, and themethod for creating this nanotip texture.A cryo-etch inductively coupled plasma isused to make the nanotips, enabling thecost-effective, mask-free formation of auniform black silicon surface over a largearea, with nanotip heights exceeding 100microns. Unlike nanotextured surfacesformed by the growth or deposition of ma-terials (e.g. carbon nanotubes), the result-ing formations are robust and compatiblewith liquid processes.This work was done by Karl Y. Yee, Eric T.

Sunada, Gani B. Ganapathi, HarishManohara, and Andrew Homyk of Caltech,and Mauro Prina of SpaceX for NASA’s JetPropulsion Laboratory. Further information iscontained in a TSP (see page 1). NPO-47299


SEM photo showing black silicon formed by ICP cryo etching.

120.10 µm

120.10µm 122.38µm

JPL LEI 5.0kV X370 10µm WD 13.6mm

Robust Multivariable Optimization and Performance Simulationfor ASIC Design Systematic method automates the simulation and optimization of circuits. Goddard Space Flight Center, Greenbelt, Maryland

Application-specific-integrated-cir-cuit (ASIC) design for space applica-tions involves multiple challenges ofmaximizing performance, minimizingpower, and ensuring reliable operationin extreme environments. This is acomplex multidimensional optimiza-tion problem, which must be solvedearly in the development cycle of a sys-tem due to the time required for test-ing and qualification severely limitingopportunities to modify and iterate.Manual design techniques, which gen-erally involve simulation at one or asmall number of corners with a verylimited set of simultaneously variableparameters in order to make the prob-lem tractable, are inefficient and notguaranteed to achieve the best possi-ble results within the performance en-velope defined by the process and en-vironmental requirements. What isrequired is a means to automate de-sign parameter variation, allow the de-signer to specify operational con-straints and performance goals, and toanalyze the results in a way that facili-tates identifying the tradeoffs defining

the performance envelope over thefull set of process and environmentalcorner cases.The system developed by the Mixed

Signal ASIC Group (MSAG) at theGoddard Space Flight Center is imple-mented as a framework of softwaremodules, templates, and function li-braries. It integrates CAD tools and amathematical computing environ-ment, and can be customized for newcircuit designs with only a modestamount of effort as most commontasks are already encapsulated. Cus-tomization is required for simulationtest benches to determine perform-ance metrics and for cost functioncomputation. Templates provide astarting point for both, while toolboxfunctions minimize the code required.Once a test bench has been coded tooptimize a particular circuit, it is alsoused to verify the final design. Thecombination of test bench and costfunction can then serve as a templatefor similar circuits or be re-used to mi-grate the design to different processesby re-running it with the new process-

specific device models. The system hasbeen used in the design of time- to-dig-ital converters for laser ranging andtime-of-flight mass spectrometry to op-timize analog, mixed signal and digitalcircuits such as charge sensitive ampli-fiers, comparators, delay elements, ra-diation tolerant dual interlocked(DICE) flip-flops, and two of threevoter gates.The overall structure of the framework

consists of two processes running inde-pendently and communicating with eachother via a data and handshaking file in-terface. The master optimization processcontains the multivariable optimizationalgorithm and performs top level analy-sis of results to select the best solution. Itis currently implemented in the Scilabopen source environment for numericalcomputation. This environment is sup-ported on multiple platforms, is wellsuited to performing numerical analysisand data visualization, and has the ad-vantage of being free to use. Cost can bea significant consideration since multiplecopies of the environment may need torun for hours or even days at a time.


The simulation process acts as a slaveprocess, simulating the circuit beingoptimized using the parameters speci-fied by the optimization process and re-

turning the results. It is currently im-plemented in the Skill programminglanguage running in the Cadence envi-ronment.

This work was done by Jeffrey DuMonthierand George Suarez of Goddard Space FlightCenter. Further information is contained in aTSP (see page 1). GSC-16185-1


Manufacturing & Prototyping

The use of mirror assemblies is com-monplace in the aerospace industry, asmost satellites and spacecraft containoptics. The fabrication of these mirrorsis extremely complex due to the natureof their intended use, which includes tel-escope lenses, camera optics, or lasermirrors. Some of the common require-ments of spacecraft mirrors are that theyhave the correct optical curvature tosome defined tolerance, they have lowsurface roughness, they have high reflec-

tivity, and they are rigid (either againstthermal expansion or flexing).Typical spacecraft mirrors are either

fabricated by coating oxide glass with ametal layer or by machining, polishing,and coating metal mirrors. Oxide glassexhibits a low coefficient of thermal ex-pansion (CTE) and can be made verysmooth but is also dense, brittle, and dif-ficult to bond to mirror-mounts. Metalmirrors are tough and light, but musttypically be diamond-turned to achieve

an optical surface and exhibit very highCTEs. Moreover, support structures,such as isogrids, flexures, or tabs, mustbe machined out of large blocks of metalor bolted on separately to the mirror.The use of different CTE materials in amirror and mirror assembly causes mis-alignment and thermal cracking.A revolutionary way to produce a mir-

ror and mirror assembly is to cast the en-tire part at once from a metal alloy thatcombines all of the desired features intothe final part: optical smoothness, curva-ture, flexures, tabs, isogrids, low CTE,and toughness. In this work, it has beendemonstrated that castable mirrors arepossible using bulk metallic glasses(BMGs, also called amorphous metals)and BMG matrix composites(BMGMCs). These novel alloys have allof the desired mechanical and thermalproperties to fabricate an entire mirrorassembly without machining, bonding,brazing, welding, or epoxy. BMGs aremulti-component metal alloys that havebeen cooled in such a manner as toavoid crystallization, leading to an amor-phous (non-crystalline) microstructure.This lack of crystal structure, and thefact that these alloys are glasses, leads toa wide assortment of mechanical andthermal properties that are unlike thoseobserved in crystalline metals. Amongthese are high yield strength, carbide-like hardness, low melting temperatures(making them castable like aluminum),a thermoplastic processing region (forimproving smoothness), low stiffness,high strength-to-weight ratios, relativelylow CTE, density similar to titanium al-loys, high elasticity, and ultra-smoothcast parts (as low as 0.2-nm surfaceroughness has been demonstrated incast BMGs). BMGMCs are composite al-loys that consist of a BMG matrix withcrystalline dendrites embeddedthroughout. BMGMCs are used to over-come the typically brittle failure ob-served in monolithic BMGs by adding asoft phase that arrests the formation ofcracks in the BMG matrix. In somecases, BMGMCs offer superior castabil-

Castable Amorphous Metal Mirrors and Mirror AssembliesCommercial applications include optics for spacecraft and satellites, mirror components fortelescopes, and mirrors for lasers, sensing, and solar energy collection.NASA’s Jet Propulsion Laboratory, Pasadena, California

Figure (a) Example of an Isogrid Mirror Backing machined from a solid block of metal. (b) An alu-minum isogrid machined from a solid plate. (c) The processing steps necessary to fabricate an amor-phous metal mirror and isogrid in a single step. An ingot of amorphous metal or composite is heatedusing radio frequency heating while sitting on water-cooled molds. The liquid is forged into the moldand cools into a glass, which allows it to be ejected from the mold without sticking. With sufficientcasting pressure, the isogrid and the mirror surface can be fabricated in one step. (d-g) The watercooled brass molds used to make the mirror, the isogrid, and the part after being ejected from themold. (h) Example of a mirror assembly with isogrid backing all made from an amorphous metal. (i) Amirror polished steel mold with curvature acts as the template for casting a perfect metallic glass mir-ror (shown in the inset).

a

b

c

1 2 3

4 5 6

d e h

f

g

i


ity, toughness, and fatigue resistance, ifnot as good a surface finish as BMGs.As shown in the figure, this work has

demonstrated that BMGs and BMGMCscan be cast into prototype mirrors andmirror assemblies without difficulty. Opti-cal curvature, ultra-smooth surfaces, andisogrids have all been demonstrated usingthis technology. A commercially manufac-tured version of these alloys would exhibita smooth surface directly off a mold, anydesired curvature or size, and all flexures,tabs, and isogrids built into the final part.This would eliminate the need for ma-chining, would reduce the CTE mismatchin the part by making it all from the samematerial and removing connections,would eliminate fatigue or stress cracking,and would be low-cost (since thousandscould be fabricated using a single mold).The novelty of the current work is that

tremendous cost and time savings can beachieved by casting the mirror assemblyinto a net or near-net shape using ametal alloy. This work suggests that mir-ror assemblies can be cast or fabricatedusing BMGs or BMGMCs where the en-tire part can be made using a simple andrepeatable processing procedure toform the mirror into the same shape as

the conventional mirror assembly, butwithout the need for machining from abillet. This idea is novel for a number ofreasons, including that conventionalmetal alloys cannot be cast into netshapes without great expense, and thisunique class of metal alloys has the abil-ity to be cast, repeatedly, into reusablemolds to net shapes and yet still have thehardness and toughness to satisfy the re-quirements for being used as a mirrorassembly. Unlike crystalline metals,BMGMCs also have unique joining prop-erties that allow them to be welded to-gether into solid structures without heat-affected zones, which removes bonding,bolting, and brazing from assembly.This work has the potential to make a

broad impact in the way that mirror as-semblies are fabricated for spacecraft,satellites, and terrestrial optics. Sincemost NASA spacecraft and satellites re-quire optics, the current invention mayprovide a path for creating high-per-formance mirrors while greatly reducingtheir cost. Moreover, the invention mayresult in an assembly-line-type processfor fabricating mirror assemblies thatwill reduce the cost of terrestrial optics,such as mirrors and telescopes.

The last technique demonstrated isa localized surface treatment as a wayto take a near-net shape and turn itinto a mirror finish. In this technique,the BMG mirror assembly is fabricatedthrough one of the strategies de-scribed above, but the mirror finish isleft in a rough state. The mirror assem-bly is then subjected to a surface treat-ment that produces an optical mirrorsurface without machining, grinding,or polishing.This work was done by Douglas C. Hof-

mann, Gregory L. Davis, Gregory S. Agnes,and Andrew A. Shapiro of Caltech for NASA’sJet Propulsion Laboratory. For more informa-tion, contact [email protected] 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 321-1234800 Oak Grove DrivePasadena, CA 91109-8099E-mail: [email protected] to NPO-48438, volume, and number

of this NASA Tech Briefs issue, and thepage number.


Mechanics/Machinery

Next-generation heat-pipe radiatortechnologies are being developed at theNASA Glenn Research Center to pro-vide advancements in heat-rejection sys-tems for space power and propulsion sys-tems. All spacecraft power andpropulsion systems require their wasteheat to be rejected to space in order tofunction at their desired design condi-tions. The thermal efficiency of theseheat-rejection systems, balanced withstructural requirements, directly affectthe total mass of the system.In testing space radiators for nuclear

power systems, it was found that cur-rent technology had significant ther-mal losses due to the long chains ofthermal resistance, stemming frommultiple materials, their respectivebonds, and physical geometries. Tubu-lar heat pipes connected to flat fins,typically made of different materials,have traditionally been thought of asthe best way to drive down radiator sys-tem mass. Dissimilar materials inher-ently had coefficient of thermal expan-sion mismatches, which createdcomplex designs and additional mass.Combining structural and thermal

components of a heat-pipe radiatorinto a unique design may produce ad-vantages to both heat transfer charac-teristics and structural integrity. TheSandwich Core Heat Pipe (SCHP) tech-nology being developed at the GlennResearch Center addresses this by usingthin titanium sheets in a unique struc-tural configuration to make a highly ef-ficient rectangular heat-pipe array thatcan be used as a thermal-control radia-tor. The concept is the first of its kind tocombine heat pipes, radiator, and struc-tural components into one system usinga single material of construction. Al-though numerous concepts are underdevelopment, the basic design allowsthe internal sandwich core structure todouble as the heat pipe vapor space, al-lowing the radiator facesheet to reachnear-vapor temperatures without the

losses associated with a conduction fin,common to current state-of-the-arttechnology. This “no fin” feature re-duces the overall temperature drop as-sociated with current heat-pipe radiatordesigns and ultimately reduces un-wanted mass through improved ther-mal efficiency. Additionally, the config-uration of the internal core using thinmetal construction provides rigidity aswell as a reduction in mass over currenttechnologies. The combination of thereduced mass and increased thermal ef-ficiency are appealing to power andpropulsion designers needing im-proved performance.Although the materials of choice for

current fission power systems are tita-nium and water, many different heat-pipe configurations could be devel-oped using other metals and working

fluids. The significance of this ap-proach is that the same basic designcould be used for multiple tempera-ture ranges, dependent only on thematerial of construction, working fluidthermophysical properties, and the ma-terial/fluid compatibility.Terrestrially, this technology could be

used for thermal control of structuralsystems. One potential use is radiantheating systems for residential andcommercial applications. The thincross section and efficient heat trans-portability could easily be applied toflooring and wall structures that couldevenly heat large surface areas. Usingthis heat-pipe technology, the evapora-tor of the radiators could be heatedusing any household heat source (elec-tric, gas, etc.), which would vaporizethe internal working fluid and carry the

Sandwich Core Heat-Pipe Radiator for Power and Propulsion SystemsThis technology has a potential use for residential or commercial radiant heating systems usinglower-cost materials of construction.John H. Glenn Research Center, Cleveland, Ohio

The Titanium Radiator incorporates multiple rectangular heat pipe channels, and an enclosed pressureboundary using a single facesheet, all from the same material. Each channel can be shared with neigh-boring channels to create a multi-channel heat pipe.

Face Sheet

Fill Tube

100.00”

Heat PipeSeparator

ChannelWall

3.00

.27

60.00”


heat to the condenser sections (wallsand/or floors). The temperature couldbe easily controlled, providing a com-fortable and affordable living environ-ment. Investigating the appropriate ma-terials and working fluids is needed to

determine this application’s potentialsuccess and usage.This work was done by Marc Gibson, James

Sanzi, and Ivan Locci of Glenn ResearchCenter. Further information is contained in aTSP (see page 1).

Inquiries concerning rights for the commer-cial use of this invention should be addressedto NASA Glenn Research Center, InnovativePartnerships Office, Attn: Steven Fedor, MailStop 4–8, 21000 Brookpark Road, Cleve-land, Ohio 44135. Refer to LEW-18900-1.

A fluid pump has been developedfor mechanically pumped fluid loopsfor spacecraft thermal control. Lyn-ntech’s technology utilizes a propri-etary electrochemically driven pump-ing mechanism. Conventionalrotodynamic and displacement pumpstypically do not meet the stringentpower and operational reliability re-quirements of space applications. Lyn-ntech’s developmental pump is ahighly efficient solid-state pump withessentially no rotating or moving com-ponents (apart from metal bellows).Figure 1 schematically illustrates the

electrochemically-driven actuator. Theconversion of electrical energy to me-chanical work is achieved by transport-ing hydrogen across an electrochemi-cal cell. Hydrogen is dissociated at theanode; protons that are formed fromthe dissociation are driven across themembrane by an applied potential,while electrons are conducted throughan external circuit; protons and elec-trons recombine at the cathode toform hydrogen. The transport of hy-drogen into and out of the attachedbellows results in a pressure variationthat is used to actuate the fluid pump-ing diaphragms. Each electrochemicalcell is made up of a proton exchangemembrane, typically Nafion®, withplatinum catalyzed electrodes on ei-ther side, called a membrane elec-trode assembly (MEA). The use of hy-drogen and its low oxidation andreduction potentials results in highelectric-to-mechanical work conver-sion efficiency. For size and convenience, a plurality

of individual MEAs is used to transportthe hydrogen, rather than a single elec-trochemical cell. The MEAs are con-nected electrically in series and fluidi-cally in parallel. The stack of MEAs issandwiched between two current collec-tors, and a voltage is applied across thestack, driving hydrogen gas from one

Apparatus for Pumping a FluidThere are no rotating or moving parts apart from bellows.Marshall Space Flight Center, Alabama

Figure 1. Hydrogen is dissociated at the anode; protons formed from the dissociation are conductedthrough the membrane, while electrons are conducted through an external circuit; protons and elec-trons recombine at the cathode to form hydrogen.

AnodeCurrent Collector

ElectrochemicalCell Cathode

Current Collector

H2 H2

H+(H2O)n

e-Anodic Reaction Cathodic Reaction

H2 2H+ 2e- 2H+ 2e- H2

++

Figure 2. The Smaller Hydrogen-Filled Bellows actuates the second larger set of bellows, which dis-place fluid in the pump heads. Fluid is expelled from one pump head while being drawn into theother.

Cooling fluid

Fluid pumpbellows

Electro-chemicalpumpactuator

H2 bellows

Cooling fluid


bellows to the other. The hydrogen flowrate through the actuator is directly pro-portional to the applied current. The hy-drogen-filled bellows are used to actuatea second set of bellows, which displacethe fluid in the pump head, as shownschematically in Figure 2. To further re-duce the pump power requirements, astroke-volume multiplier is utilizedwherein a smaller-volume hydrogenfilled bellows actuates a larger-volumefluid bellows. The stroke volume multi-plier also allows the pump frequency tobe reduced below audible frequencywhile maintaining adequate flow.

The largest factor affecting the lifetimeand reliability of the pump is expected tobe loss of hydrogen from the electro-chemical actuator. The electrochemicalactuator is hermetically sealed; however,permeation of hydrogen is expected toeventually result in loss of hydrogen. Thelifetime of the pump is extended by gen-erating hydrogen onboard the pump.The onboard hydrogen generation alsoallows the hydrogen pressure and pumpperformance to be optimized for varyingoperating temperatures. The prototype pump is expected to op-

erate with a power consumption of 2.4 W

at a flow rate of 0.76 L/min and pressurerise of 27.6 kPa. The pump will operate attemperatures between 0 and 100 °C andsurvive temperatures between –60 to 110°C. The prototype occupies a volume of≈600 cm3.This work was done by Robert Van Boeyen

and Jonathan Reeh of Lynntech, Inc. forMarshall Space Flight Center. For more infor-mation, contact Sammy A. Nabors, MSFCCommercialization Assistance Lead, [email protected]. Refer to MFS-32760-1.

A prime focus spectrometer (PFS),along with corrective optics, will mountin place of the secondary mirror of theSubaru telescope on Mauna Kea,Hawaii. This will allow simultaneous ob-servations of cosmologic targets. It willenable large-scale galactic archeologyand dark energy surveys to help unlockthe secrets of the universe.To perform these cosmologic surveys,

an array of 2,400 optical fibers needs tobe independently positioned within the498-mm-diameter focal plane of the PFSinstrument to collect light from galaxiesand stars for spectrographic analyses. Toallow for independent re-positioning ofthe fibers, a very small positioner (7.7mm in diameter) is required. One hun-dred percent coverage of the focalplane is also required, so these small ac-tuators need to cover a patrol region of9.5 mm in diameter. To optimize theamount of light that can be collected,the fibers need to be placed within 5 mi-crometers of their intended target (ei-ther a star or galaxy).The Cobra Fiber Positioner was de-

signed to meet the size and accuracy re-quirements stated above. Cobra is a two-degrees-of-freedom mechanism that canposition an optical fiber in the focal

plane of the PFS instrument to a preci-sion of 5 micrometers. It is a theta-phistyle positioner containing two rotarypiezo tube motors with one offset fromthe other, which enables the optic fibersto be placed anywhere in a small circularpatrol region. The patrol region of theactuator is such that the array of 2,400positioners allows for full coverage ofthe instrument focal plane by overlap-ping the patrol areas.A second-generation Cobra posi-

tioner was designed based on lessonslearned from the original prototype

built in 2009. Improvements were madeto the precision of the ceramic motorparts, and hard stops were redesignedto minimize friction and prevent jam-ming. These changes resulted in reduc-ing the number of move iterations re-quired to position the optical fiberwithin 5 micrometers of its target. At thetime of this reporting, there are stillmany tests to be performed that will val-idate system level performance, but onan individual level, the Cobra positionerdemonstrates excellent performanceand will enable the PFS instrument tomake unprecedented measurements ofthe universe.What is unique about the upgrades

made to the Cobra positioner is the im-proved performance due to the designchanges in the hard stops and the ce-ramic end caps of the motors. Otherchanges were made to reduce the unitcost of a Cobra positioner without affect-ing the performance, since thousands ofthese devices will have to be built for thePFS instrument.This work was done by Charles D. Fisher,

David F. Braun, and Joel V. Kaluzny of Cal-tech for NASA’s Jet Propulsion Laboratory.For more information, contact [email protected]. NPO-48751

Cobra Fiber-Optic Positioner UpgradeThis technology could be used for applications requiring precise location of a small objectwithin a small circular area, such as in medical lasers.NASA’s Jet Propulsion Laboratory, Pasadena, California

The Prime Focus Spectrometer (PFS) installed onthe Subaru Telescope.


Materials & Coatings

The objective of this effort was todevelop a non-toxic, non-flammable,–80 °C phase change material (PCM)to be used in NASA’s ICEPAC cap-sules for biological sample preserva-

tion in flight to and from Earth orbit.A temperature of about –68 °C orlower is a critical temperature formaintaining stable cell, tissue, andcell fragment storage.

Within this technical effort, two phasechange fluids were developed with melt-ing onset at –85 °C and –61 °C, and la-tent heat of fusion of 100 and 136 J/mL,respectively. The experimental results

Non-Toxic, Non-Flammable, –80 ºC Phase Change MaterialsLyndon B. Johnson Space Center, Houston, Texas

Improved Wide Operating Temperature Range of Li-Ion Cells Applications include electric vehicles, where high-energy-density and high-power batteries areneeded that can operate at low temperature.NASA’s Jet Propulsion Laboratory, Pasadena, California

Future NASA missions aimed at ex-ploring the Moon, Mars, and the outerplanets require rechargeable batteriesthat can operate over a wide temperaturerange (–60 to +60 ºC) to satisfy the re-quirements of various applications in-cluding landers, rovers, penetrators,CEV, CLV, etc. This work addresses theneed for robust rechargeable batteriesthat can operate well over a wide temper-ature range.The Department of Energy (DoE) has

identified a number of technical barriersassociated with the development of Li-ion rechargeable batteries for PHEVs.For this reason, DoE has interest in thedevelopment of advanced electrolytesthat will improve performance over awide range of temperatures, and lead tolong life characteristics (5,000 cycles overa 10-year life span). There is also interestin improving the high-voltage stability ofthese candidate electrolyte systems to en-able the operation of up to 5 V with highspecific energy cathode materials.Currently, the state-of-the-art lithium-ion

system has been demonstrated to operateover a wide range of temperatures (–40 to+40 ºC); however, the rate capability at thelower temperatures is very poor. In addi-tion, the low-temperature performancetypically deteriorates rapidly upon beingexposed to high temperatures.A number of electrolyte formulations

were developed that incorporate the useof electrolyte additives to improve thehigh-temperature resilience, low-tempera-

ture power capability, and life characteris-tics of methyl propionate (MP)-basedelectrolyte solutions. These electrolyte ad-ditives include mono-fluoroethylene car-bonate (FEC), lithium oxalate, vinylenecarbonate (VC), and lithium bis(oxalateborate) (LiBOB), which have previouslybeen shown to result in improved high-temperature resilience of all carbonate-based electrolytes. These MP-based elec-trolytes with additives have been shown tohave improved performance in experi-ments with MCMB-LiNiCoAlO2 cells.A number of lithium-ion electrolytes

having improved temperature range ofoperation were demonstrated. LiPF6-based mixed carbonate electrolyte for-mulations that contain ester co-solventshave been optimized for operation atlow temperature, while still providingreasonable performance at high temper-ature. In earlier work [see “OptimizedCarbonate and Ester-Based Li-Ion Elec-trolytes” (NPO-44974) NASA Tech Briefs,Vol. 32, No. 4 (April 2008), p. 56], esterco-solvents, including methyl propi-onate (MP), ethyl propionate (EP),methyl butyrate (MB), ethyl butyrate(EB), propyl butyrate (PB), and butylbutyrate (BB), were investigated inmulti-component electrolytes of the fol-lowing composition: 1.0 M LiPF6 in eth-ylene carbonate (EC) + ethyl methyl car-bonate (EMC) + X (20:60:20 v/v %)[where X = ester co-solvent]. Focusingupon improved rate capability at lowtemperatures (i.e., –20 to –40 ºC), this

approach was optimized further [see“Li-Ion Cells Employing ElectrolytesWith Methyl Propionate and Ethyl Bu-tyrate Co-Solvents” (NPO-46976), NASATech Briefs, Vol. 35, No. 10 (October2011), p. 47], which resulted in the de-velopment of 1.20M LiPF6 inEC+EMC+MP (20:20:60 v/v %) and1.20M LiPF6 in EC+EMC+EB (20:20:60v/v %), which were demonstrated to op-erate well over a wide temperature rangein MCMB-LiNiCoA1O2 and Li4Ti5O12-LiNiCoA1O2 prototype cells. In the cur-rent work, improved high temperatureresilience, low temperature power capa-bility, and life characteristics have beenprovided with methyl propionate-basedelectrolyte solutions [i.e., 1.20M LiPF6 inEC+EMC+MP (20:20:60 v/v%)] possess-ing the additives described above.This work was done by Marshall C. Smart

and Ratnakumar V. Bugga of Caltech forNASA’s Jet Propulsion Laboratory. Further in-formation 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 321-1234800 Oak Grove DrivePasadena, CA 91109-8099E-mail: [email protected] to NPO-47538, volume and number



indicate good repeatability of thefreeze/thaw cycle, compatibility withhigh-density polyethylene, thermal sta-bility, and flashpoints exceeding 100 °C.Based on the individual components,the phase change fluids are expected tohave low acute toxicity. There are several types of phase

change materials (PCMs) that can beconsidered for the preservation of biolog-ical samples at a temperature of about–68 ºC. These include hydrated salts inwater, non-hydrated or weakly hydratedsalts in water, non-electrolyte aqueous so-lutions, and pure, non-aqueous fluids.However, none of these options resulted

in a –80 °C freezing point stable PCMduring the freeze/thaw cycling. There-fore, a non-aqueous mixture was formu-lated yielding a pseudo melting pointplateau, adequate stability over numer-ous thermal cycles, and suitable latentheats. The product is a non-aqueous basefluid with a somewhat tailored freezingpoint, and latent heats on the order of100 J/mL for the –85 °C PCM.The fluid developed is an organic so-

lution with adequate resistance to bio-logical growth, compatible with HDPE(high density polyethylene) plastic,and is characterized by a negativefreezing expansion ratio. The negative

expansion ratio during freezing willallow for the ICEPAC modules to becompletely filled with PCM material ascompared to previously used fluids re-quiring a 20-mL air bubble (within a120-mL capsule). This translates to a20-percent increase in cooling capacityfor a given latent heat. Furthermore,the specific gravity of the PCM is on theorder of 0.92 g/mL, making it lighterthan an aqueous-based solution perICEPAC capsule.This work was done by J. Michael Cutbirth

of Mainstream Engineering Corp. for JohnsonSpace Center. Further information is con-tained in a TSP (see page 1). MSC-24460-1

The “soft-bake” method is a simpleand reliable initial purification step firstproposed by researchers at Rice Univer-sity for single-walled carbon nanotubes(SWCNT) produced by high-pressurecarbon mon oxide disproportionation(HiPco). Soft-baking consists of anneal-ing as-produced (raw) SWCNT, at lowtemperatures in humid air, in order todegrade the heavy graphitic shells thatsurround metal particle impurities.Once these shells are cracked open bythe expansion and slow oxidation of themetal particles, the metal impurities canbe digested through treatment with hy-drochloric acid. The soft-baking of SWCNT produced

by pulsed-laser vaporization (PLV) is notstraightforward, because the larger aver-age SWCNT diameters (≈1.4 nm) andheavier graphitic shells surroundingmetal particles call for increased temper-atures during soft-bake. A part of thetechnology development focused on op-timizing the temperature so that effec-tive cracking of the graphitic shells isbalanced with maintaining a reasonableyield, which was a critical aspect of thisstudy. Once the ideal temperature wasdetermined, a number of samples of rawSWCNT were purified using the soft-bake method.An important benefit to this process is

the reduced time and effort required forsoft-bake versus the standard purifica-tion route for SWCNT. The total time

spent purifying samples by soft-bake isone week per batch, which equates to afactor of three reduction in the time re-quired for purification as compared tothe standard acid purification method.Reduction of the number of steps alsoappears to be an important factor in im-proving reproducibility of yield and pu-rity of SWCNT, as small deviations arelikely to get amplified over the course ofa complicated multi-step purificationprocess. The full JSC characterization proto-

col consisting of UV-Vis-NIR absorption,Raman spectroscopy, thermogravimet-ric analysis (TGA), scanning electronmicroscopy (SEM), and transmissionelectron microscopy (TEM) was appliedto all samples. In addition, the nan-otube content as the percentage out oftotal carbon was calculated from UV-Vis-NIR absorption, using a procedure sim-ilar to that proposed in earlier research.The nanotube percentage relative tothe total sample weight was calculatedfrom the absorption data, taking intoaccount the metal content. These meas-urements were used to quantify variousaspects of the “quality” of the materialin terms of: metal content (TGA residueweight percentage), oxidation tempera-ture (the temperature of the most sig-nificant peak in the first derivative ofthe TGA), stability of DMF dispersion byoptical absorption, nanotube content(weight percentage of total from UV-

Vis-NIR absorption and TGA), the fre-quency of the Raman G-band peak, andthe ratio between the D-band and G-band intensities. All measurements (ex-cept dispersion stability) were repeatedthree times to determine variability ofthe purification route through statisticalmethods. Standard deviations of the ex-perimental results are considered im-portant parameters to quantify the ho-mogeneity of nanotube samples. The data indicate that the major

properties, such as metal content, nan-otube content, oxidation temperature,and extent of defects as determined byRaman, are similar or superior for thesoft-baked samples when compared tothe standard acid-purified samples.TGA and Raman data suggest that thereis more metal removed, less oxidativedamage, less protonation, and less de-rivatization incurred from the acidtreatment in soft-baked samples. Thestability and reproducibility of DMF dis-persions is also superior, suggesting thatsoft-baking leads to greater homogene-ity in materials than standard acid pu-rification routes.The observed improvement in effi-

ciency translates directly into greateryield and quality of nanotubes, re-duced cost and processing time, andthe use of lesser amounts of organicsolvents and concentrated acids, mak-ing the process safer and more envi-ronmentally friendly. This approach, as

Soft-Bake Purification of SWCNTs Produced by Pulsed Laser VaporizationA more efficient, cost-effective, environmentally friendly method purifies high-quality carbon nanotubes.Lyndon B. Johnson Space Center, Houston, Texas


well as the development of a qualitycontrol, may be useful in the industrialscale-up of production and purificationof carbon nanotubes.

This work was done by Leonard Yowell ofJohnson Space Center, and Pavel Nikolaev,Olga Gorelik, Rama Kumar Allada, EdwardSosa, and Sivaram Arepalli of ERC, Inc. Fur-

ther information is contained in a TSP (seepage 1). MSC-24379-1


An improved method for culturing im-mature muscle cells (myoblasts) into amature skeletal muscle overcomes someof the notable limitations of prior culturemethods. The development of themethod is a major advance in tissue engi-neering in that, for the first time, a cell-based model spontaneously fuses and dif-ferentiates into masses of highly aligned,contracting myotubes. This method en-ables (1) the construction of improvedtwo-dimensional (monolayer) skeletalmuscle test beds; (2) development ofcontracting three-dimensional tissuemodels; and (3) improved transplantabletissues for biomedical and regenerativemedicine applications. With adaptation,this method also offers potential applica-tion for production of other tissue types(i.e., bone and cardiac) from correspon-ding precursor cells.In order to form or repair skeletal

muscle tissue, myoblasts must prolifer-ate, align, fuse to form myotubes (ma-turing, elongated, multinucleated cells),and differentiate to produce new pro-teins necessary for contraction. Hereto-fore, none of the typical culture meth-ods has enabled the growth of skeletalmuscle tissue possessing all of thesecharacteristics. Notable limitations ofprior methodologies include:• Differentiation is usually induced bythe use of undefined serum switches ordeprivation, either of which can resultin decreased cell viability, occasionalunpredictable variations, and experi-mental bias.• Alignment of cells is affected by meansof time-consuming procedures that in-volve the use of expensive, complex,micropatterned substrates.• Spontaneous contraction is a random,rare event. In contrast, the present method en-

ables the generation of large numbers ofaligned, spontaneously contracting my-otubes without the use of serum switchesor special substrates. The resources nec-essary for practicing this method arereadily available in most cell-culture lab-

oratories.The method is embodied in a cost-ef-

fective, multistep protocol of cell culturethat results in the formation of the de-sired skeletal muscle tissue model andthat, with additional time in culture,yields sheets of spontaneously contract-ing aligned muscle cells. Omitting mostdetails for the sake of brevity, the proto-col is summarized as follows:1. From typical cell culture of the de-sired cell type, the cells are removedand concentrated. The cells are resus-pended in culture medium to achievea specific cell density.

2. Cells are seeded onto the bottoms oftissue culture Petri dishes.

3. New culture medium is added to thedish and cultured at 37 °C. After 3days of incubation, myotubes are evi-dent. At 5 to 6 days, myotubes arealigned and are in greater abundance.

4. From this point onward, Petri dishesshould be monitored on a daily basisand medium should be replaced withculture medium on an as-needed basis.

5. By the end of a two-week culture pe-riod, spontaneously contracting my-otubes are evident although contrac-tions may have been noted as early asday 5.

6. Contracting myotubes can be main-tained for an additional two weeks withdaily culture medium replacement.During this additional culture time,sheets of aligned contracting myotubesappear to gain synchronicity.The theory underlying this methodol-

ogy includes the following elements:• Low-density cell culture favors cell flat-tening. Flattened cell morphology, inthe presence of growth factors found inthe culture medium, encourages cellproliferation. In this new methodology,the lower-density cell population willalso proliferate. The newly formed cells,crowded by adjacent newly formedcells, are forced to compete for attach-ment space and are forced to align.• Simultaneously, the cell crowdingcaused by the high-density cell popula-tion prevents total cell flattening. In-

Improved Cell Culture Method for Growing ContractingSkeletal Muscle ModelsThis method has great potential for biomedical research and medical treatment.Lyndon B. Johnson Space Center, Houston, Texas

Bio-Medical

Comparison of Skeletal Muscle Culture Methods: Typical culture using serum switches results in ran-dom distribution of myotubes (left). The arrow indicates one of many myotubes in the image. The im-proved cell culture methodology produces aligned myotubes (right) without the use of (micropat-terned) substrate modification.


stead, the attached, rounded morphol-ogy, coupled with the high degree andduration of cell-to-cell contact, isthought to induce expression of genesthat causes the cells to switch from pro-liferating to differentiating phenotype.This differentiating phenotype gainsthe ability to fuse and express contrac-tile proteins.Currently, this model can serve as an

improved, yet easily generated, in vitroskeletal muscle test bed for evaluatingsignal transduction pathways, stretch ac-tivated ion channels, protein synthesisregulation, and interactions with the ex-

tracellular matrix and membrane pro-teins with the cytoskeleton. The easeand low cost of generation makes thismodel a valuable option to many labora-tories that have the desire, but lack thesophisticated resources previously re-quired to explore the mechanisms ofmuscle atrophy. Future development ofthe model into a three-dimensional con-struct provides a potential powerful toolfor defining the mechanisms of micro-gravity-induced muscle atrophy, identifi-cation of molecular targets for novelcountermeasure development, and test-ing of some countermeasures.

This work was done by Michele L. Mar-quette of the University of Texas MedicalBranch, and Marguerite A. Sognier of theUniversities Space Research Association forJohnson Space Center. For further informa-tion, contact the JSC Innovation PartnershipsOffice at (281) 483-3809.This invention is owned by NASA, and a

patent application has been filed. Inquiriesconcerning nonexclusive or exclusive licensefor its commercial development should be ad-dressed to the Patent Counsel, Johnson SpaceCenter, (281) 483-1003. Refer to MSC-24314-1

Hand-Based Biometric AnalysisGoddard Space Flight Center, Greenbelt, Maryland

Hand-based biometric analysis systemsand techniques provide robust hand-based identification and verification. Animage of a hand is obtained, which isthen segmented into a palm region andseparate finger regions. Acquisition ofthe image is performed without requir-ing particular orientation or placementrestrictions. Segmentation is performedwithout the use of reference points on

the images. Each segment is analyzed bycalculating a set of Zernike moment de-scriptors for the segment.The feature parameters thus obtained are

then fused and compared to stored sets ofdescriptors in enrollment templates to arriveat an identity decision. By using Zernike mo-ments, and through additional manipula-tion, the biometric analysis is invariant to ro-tation, scale, or translation or an input

image. Additionally, the analysis uses re-useof commonly seen terms in Zernike calcula-tions to achieve additional efficiencies overtraditional Zernike moment calculation. This work was done by George Bebis of Uni-

versity of Nevada, Reno for Goddard SpaceFlight Center. For further information, contactthe Goddard Innovative Partnerships Office at(301) 286-5810. GSC-16141-1

The Next Generation of Cold Immersion Dry Suit DesignEvolution for Hypothermia Prevention The system design recovers warm exhaled air and re-circulates it inside the suit. John H. Glenn Research Center, Cleveland, Ohio

A body at sea is vulnerable to hy-pothermia, which often leads to loss oflife. Hypothermia is caused by the dif-ferences between the core body tem-perature and the surrounding air andseawater temperatures. The greater thedifferences between the body core tem-perature and the sea temperature, themore rapidly the core body tempera-ture will drop, and hypothermia canquickly set in. Heat loss is primarilycaused by conduction of heat awayfrom the body. Most cold immersionsuits on the market are passive designsthat only insulate the body against thecold, although some cold immersionsuits use special materials such as paraf-fin to absorb heat and to radiate theheat back to the body. This new utilitypatent is an active design that relies onthe lung’s role as an organic heat ex-

changer for providing deep body coreheating of air. It is based on the factthat the greatest heat loss mechanismfor an insulated human body immersedin a cold water environment is due toheat loss through respiration. This innovation successfully merges

two existing technologies (cold immer-sion suit and existing valve technolo-gies) to produce a new product thathelps prevent against the onset of hy-pothermia at sea. During normal opera-tions, a human maintains an approxi-mate body temperature of [98.6 °F (37°C)]. A mechanism was developed to re-cover the warm temperature from thebody and reticulate it in a survival suit.The primary intention is to develop anencompassing systems design that canboth easily and cost effectively be inte-grated in all existing currently manufac-

tured cold water survival suits, and assuch, it should be noted that the coldwater immersion suit is only used as aframework or tool for laying out the re-quired design elements. At the heart of the suit is the Warm Air

Recovery (WAR) system, which relies ona single, large Main Purge Valve (MPV)and secondary Purge Valves (PV) to op-erate. The main purge valve has a thinmembrane, which is normally closed,and acts as a one-way check valve. Whenwarm air is expelled from the lungs, itcauses the main purge valve to open. Airforced from the MPV is dumped directlyinto the suit, thereby providing warmthto the torso, legs, and arms. A slight pos-itive over-pressure in the suit causeswarm waste air (or water if the suit ispunctured) to be safely vented into thesea through large PVs located at the bot-


tom of each arm and leg. The secondarypurge valves act to prevent the buildupof large concentrations of CO2 gas andhelp guard against asphyxia. It is notedthat the MPV causes the inhalation andexhalation cycles to be completely iso-lated from one another in the currentsuit design. The main problem with the existing

survival suit designs stems from cold seawater entering the suit through the gapsaround the face opening in the hood andinflatable neck collar. Cold water enteringthe suit is caused by high wind and waves,a problem that is further exacerbated ifthe suit is submerged by a wave. The prob-lem is easily solved using four integralcomponents: an optically transparent fullface shield, a nose/mouth mask, a Sur-face Air Valve (SAV), and a MPV.Surface air valve integration with the

suit is effected by affixing the SAV on theoutward side of the face shield with ashort tube extending through the faceshield, which is directly plumbed to theair inlet port on the mask. WAR systemintegration with the suit is achieved byconnecting the MPV to the air outletport located at the bottom of the masknear the mouth.A special type of SAV mechanism

called a Pressure Release Valve (PRV) isemployed in the current design. Whenthe PRV is above the sea surface, thePRV is open, allowing the lungs to in-

hale a fresh atmospheric air. The PRVautomatically closes when submerged.This PRV has the advantage of beingable to operate in any position, evenwhile completely submerged upsidedown. The PRV is connected to a shorttube extending through the face shield,and the open end of the tube is termi-nated with yet another purge valve lo-cated at the air inlet port on the mask.The PRV and PV combo results in aunique new type of valve called a Pres-sure Release/Purge valve (PR/P) that isdesigned to compensate for additionalheat loss from the respiration cycle andonly applies to situations when the PR/Pvalve is above the sea surface. The fol-lowing two examples are cosidered: Example (1) is a situation in which the

air inlet port on the mask is connected tothe PRV by a short tube extendingthrough the face shield. Since the PRV isabove the sea surface, the PRV is openand air can freely be drawn into thelungs. When warm air is exhaled fromthe lungs, it causes air to flow outwardthrough the PRV, causing the body tolose valuable heat during the exhalationcycle. Therefore, in this example, the in-halation and exhalation cycles are nolonger separated from one another.Example (2) is a situation in which the

PR/P valve on the face shield is con-nected to the air inlet port on the mask,and the MPV is connected to the air out-

let port on the mask. Because the PR/Pvalve is above the sea surface, the PR/Pvalve is open. The action of drawing airfrom the mask causes the PV located atthe end of the PR/P valve to open and theMPV to close. When air is exhaled fromthe lungs into the mask, it causes the PVlocated on the end of the PR/P valve toclose, and warm exhaled air flowing intothe mask forces the MPV to open, ventingwarm air directly into the suit. Thus thePR/P valve recovers any heat loss duringthe exhalation part of the respirationcycle.The new technology represents a sub-

stantial improvement over existing state-of-the-art survival suit designs and shouldgreatly extend the length of time a bodycan survive in hazardous cold-water condi-tions encountered in the North Atlantic,Bering Sea, and other cold regions of theworld. This suit technology is intended tobe utilized by the various branches of theU.S. Coast Guard, the U.S. Navy, the Mer-chant Marine, as well as the commercialfishing industry fleet.This work was done by Joel Galofaro of

Glenn Research Center. Further information iscontained in 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: Steven Fedor, MailStop 4–8, 21000 Brookpark Road, Cleveland,Ohio 44135. Refer to LEW-18960-1.


Software

Integrated Lunar Informa-tion Architecture for Deci-sion Support Version 3.0(ILIADS 3.0)

ILIADS 3.0 provides the data manage-ment capabilities to access CxP-vettedlunar data sets from the LMMP-providedData Portal and the LMMP-provided On-Moon lunar data product server. (LMMPstands for Lunar Mapping and ModelingProject.) It also provides specific quanti-tative analysis functions to meet thestated LMMP Level 3 functional and per-formance requirements specificationsthat were approved by the CxP. ILIADS 3.0 is a rich client (aka, thick

client) lunar Geospatial InformationSystem (GIS) software application. It is aredesigned software framework and ar-chitecture that leverages GSFCs experi-ence developing the ILIADS 2.0 coresoftware. Specifically, ILIADS 3.0 is builtupon ILIADS 2.0. The purpose of ILIADS 3.0 is to pro-

vide an integrated, rich client lunarGIS software application. Most signifi-cantly, the objective in designing anddeveloping ILIADS 3.0 was to provide aflexible and expandable softwareframework that readily enabled newfeatures and functions to be integratedinto ILIADS driven by the science andengineering user community. To con-tribute to the decision support process,ILIADS 3.0 also provides interfaces toreadily enable interoperability betweenILIADS 3.0 and other NASA-developedlunar information systems whenever itmay become required to interface ILI-ADS with such systems. By building upon Goddard’s IRC core

framework, ILIADS is also well suited tobeing readily integrated with futurelunar surface system assets (e.g., crewedrovers, spacesuits) as an embedded sys-tem application. ILIADS 3.0 providescross-platform support and thus exe-cutes on a diverse suite of computingplatforms that are used by NASA scien-tists and engineers. The application isdesigned to provide author ized/authenticated users with the ability touse the Internet to securely, yet easily,identify and locate geographically dis-tributed sources of vetted lunar dataproducts that have been derived fromUS and international lunar spacecraftmissions. It can query the data catalogs

of these sources to identify availablelunar data products and the metadataassociated with them. The software usesstandard OGC (Open Geospatial Con-sortium) WMS, WCS, and WFS (WebMap Service, Web Coverage Service,and Web Feature Service) services to ac-cess mapped lunar delta products fromthese sources so that they may beprocessed by ILIADS 3.0 and renderedas multiple semi-transparent raster orvector visualizations, so that the lunardata product information is readily un-derstood in the context of one another.This work was done by Stephen Talabac,

Troy Ames, Karin Blank, Carl Hostetter, andMatthew Brandt of Goddard Space FlightCenter. Further information is contained in aTSP (see page 1). GSC-16210-1

Relay Forward-Link FileManagement Services(MaROS Phase 2)

This software provides the service-level functionality to manage the deliv-ery of files from a lander mission repos-itory to an orbiter mission repositoryfor eventual spacelink relay by the or-biter asset on a specific communica-tions pass. It provides further functionsto deliver and track a set of mission-de-fined messages detailing lander author-ization instructions and orbiter data de-livery state. All of the informationconcerning these transactions is per-sisted in a database providing a highlevel of accountability of the forward-link relay process.This is an improvement over legacy

processes that required lander clientusers to log into orbiter mission worksta-tions and run orbiter-specific applica-tions. The legacy process provided onlya simple e-mail indicating success oftransaction, and no further accountingof the forward-link transaction.The Phase 2 MaROS forward-file

management functions represent a sig-nificant upgrade of this relay system.This version provides a lander team thecapability of selecting a set of “forward-link” files to be radiated to an orbiterfor relay during a chosen communica-tions window. These forward-link filescontain critical flight data such asspacecraft command sequences andflight software uploads; this newMaROS functionality is classified Class

B software. This new software versionalso includes the capability for landerand orbiter team members to associatepredefined messages to the chosen setof forward-link files. Lander teams canspecify authorizations for the orbiterteam such as “go for radiation,” and or-biter team members may specify statusmessages such as “onboard spacecraft.”All of the status data is tracked in adatabase and provided via a sharedservice interface. The system provides ahigh level of accountability into the for-ward-link process.This work was done by Daniel A. Allard,

Michael N. Wallick, Franklin H. Hy, and RoyE. Gladden of Caltech for NASA’s Jet Propul-sion Laboratory. Further information is con-tained in a TSP (see page 1).This software is available for commercial li-

censing. Please contact Dan Broderick [email protected]. Refer to NPO-47942.

Two Mechanisms to AvoidControl Conflicts Resultingfrom Uncoordinated Intent

This software implements a real-timeaccess control protocol that is intendedto make all connected users aware of thepresence of other connected users, andwhich of them is currently in control ofthe system. Here, “in control” meansthat a single user is authorized and en-abled to issue instructions to the system.The software also implements a goal

scheduling mechanism that can detectsituations where plans for the operationof a target system proposed by differentusers overlap and interact in conflictingways. In such situations, the system caneither simply report the conflict (reject-ing one goal or the entire plan), orreschedule the goals in a way that doesnot conflict.The access control mechanism (and

associated control protocol) is unique.Other access control mechanisms aregenerally intended to authenticate users,or exclude unauthorized access. Thissoftware does neither, and would likelydepend on having some other mecha-nism to support those requirements.This work was done by Andrew H.

Mishkin, Daniel L. Dvorak, David A. Wag-ner, and Matthew B. Bennett of Caltech forNASA’s Jet Propulsion Laboratory. Further in-formation is contained in a TSP (see page 1).


This software is available for commercial li-censing. Please contact Dan Broderick [email protected]. Refer to NPO-47732.

XTCE GOVSAT Tool Suite 1.0

The XTCE GOVSAT software suitecontains three tools: validation, search,and reporting. The Extensible MarkupLanguage (XML) Telemetric and Com-mand Exchange (XTCE) GOVSAT ToolSuite is written in Java for manipulating

XTCE XML files. XTCE is a Consulta-tive Committee for Space Data Systems(CCSDS) and Object Man agementGroup (OMG) specification for describ-ing the format and information intelemetry and command packetstreams. These descriptions are filesthat are used to configure real-timetelemetry and command systems formission operations. XTCE’s purpose isto exchange database information be-tween different systems. XTCE GOVSAT consists of rules for

narrowing the use of XTCE for missions.

The Validation Tool is used to syntaxcheck GOVSAT XML files. The SearchTool is used to search (i.e. commandand telemetry mnemonics) the GOV-SAT XML files and view the results. Fi-nally, the Reporting Tool is used to cre-ate command and telemetry reports.These reports can be displayed orprinted for use by the operations team.This work was done by J. Kevin Rice of

Global Science & Technology, Inc. for God-dard Space Flight Center. Further informationis contained in a TSP (see page 1). GSC-16279-1


Information Technology

SequenceL: Automated Parallel Algorithms Derived from CSP-NT Computational LawsChip manufacturers and developers of parallel and/or safety-critical software could benefit from this innovation. Goddard Space Flight Center, Greenbelt, Maryland

With the introduction of new par-allel architectures like the cell andmulticore chips from IBM, Intel,AMD, and ARM, as well as the petas-cale processing available for high-end computing, a larger number ofprogrammers will need to write par-allel codes. Adding the parallel con-trol structure to the sequence, selec-tion, and iterative control constructsincreases the complexity of code de-velopment, which often results in in-creased development costs and de-creased reliability.SequenceL is a high-level program-

ming language — that is, a program-ming language that is closer to ahuman’s way of thinking than to a ma-chine’s. Historically, high-level lan-guages have resulted in decreased devel-opment costs and increased reliability,at the expense of performance. In re-cent applications at JSC and in industry,SequenceL has demonstrated the usualadvantages of high-level programmingin terms of low cost and high reliability.

SequenceL programs, however, haverun at speeds typically comparable with,and in many cases faster than, theircounterparts written in C and C++ whenrun on single-core processors. More-over, SequenceL is able to generate par-allel executables automatically for mul-ticore hardware, gaining parallelspeedups without any extra effort fromthe programmer beyond what is re-quired to write the sequen tial/single-core code. A SequenceL-to-C++ translator has

been developed that automatically ren-ders readable multithreaded C++ from acombination of a SequenceL programand sample data input. The SequenceLlanguage is based on two fundamentalcomputational laws, Consume-Simplify-Produce (CSP) and Normalize-Trans -pose (NT), which enable it to automatethe creation of parallel algorithms fromhigh-level code that has no annotationsof parallelism whatsoever. In our anec-dotal experience, SequenceL develop-ment has been in every case less costly

than development of the same algo-rithm in sequential (that is, single-core,single process) C or C++, and an orderof magnitude less costly than develop-ment of comparable parallel code.Moreover, SequenceL not only automat-ically parallelizes the code, but since it isbased on CSP-NT, it is provably race free,thus eliminating the largest quality chal-lenge the parallelized software devel-oper faces. Compiling functional code to C++ is

not new. Compiling functional code toreadable C++ that runs in parallel ismuch more of a challenge, and that wasthe majority of this effort. For currentpurposes in this effort, readability of thegenerated code is crucial, in case thehuman programmer wishes to add anno-tations, or to inspect the code for verifi-cation purposes. Moreover, by compilingto C++ it is assured that SequenceL canbe used in any application where C++could be used. SequenceL has been found to dis-

cover all potential parallelisms automat-

Determining Temperature Differential to Prevent HardwareCross-Contamination in a Vacuum Chamber Goddard Space Flight Center, Greenbelt, Maryland

When contamination-sensitive hard-ware must be tested in a thermal vac-uum chamber, cross-contaminationfrom other hardware present in thechamber, or residue from previous tests,becomes a concern. Typical mitigationstrategies involve maintaining the tem-perature of the critical item above thatof other hardware elements at the endof the test. A formula for relating the pumping

speed of a chamber, the surface area ofcontamination sources, and the tempera-tures of the chamber, source, and con-

tamination-sensitive items has been devel-oped. The formula allows the determina-tion of a temperature threshold aboutwhich contamination will not condenseon the sensitive items. It defines a param-eter alpha that is the fraction given by(contaminant source area)/[chamberpumping speed × (time under vac-uum)0.5]. If this parameter is less than10–6, cross-contamination from commonspacecraft material will not occur whenthe sensitive hardware is at the samete,mperature as the source of contamina-tion (The chamber is isothermal within 5

°C.). Knowing when it becomes safe to have

the hardware isothermal permits fasterand easier thermal transitions whencompared with maintaining an arbitrarytemperature differential between parts.Furthermore, the standard temperaturedifferential may not be adequate undersome conditions (alpha>10–4). This work was done by David Hughes of God-

dard Space Flight Center. For further informa-tion, contact the Goddard Innovative Partner-ships Office at (301) 286-5810. GSC-16244-1


A difficulty for many NASA researchersis that often the data to analyze is locatedremotely from the scientist and the datais too large to transfer for local analysis.Researchers have developed the Data Ac-cess Protocol (DAP) for accessing remotedata. Presently one can use DAP fromwithin IDL, but the IDL-DAP interface isboth limited and cumbersome. A morepowerful and user-friendly interface toDAP for IDL has been developed. Users are able to browse remote data

sets graphically, select partial data to re-trieve, import that data and make cus-tomized plots, and have an interactive

IDL command line session simultaneouswith the remote visualization. All ofthese IDL-DAP tools are usable easilyand seamlessly for any IDL user. IDL and DAP are both widely used in

science, but were not easily used to-gether. The IDL DAP bindings were in-complete and had numerous bugs thatprevented their serious use. For exam-ple, the existing bindings did not readDAP Grid data, which is the organiza-tion of nearly all NASA datasets cur-rently served via DAP. This project uniquely provides a fully

featured, user-friendly interface to DAP

from IDL, both from the command lineand a GUI application. The DAP Ex-plorer GUI application makes browsinga dataset more user-friendly, while alsoproviding the capability to run user-de-fined functions on specified data. Meth-ods for running remote functions on theDAP server were investigated, and atechnique for accomplishing this taskwas decided upon. This work was done by Michael Galloy of

Tech-X Corporation for Goddard Space FlightCenter. Further information is contained in aTSP (see page 1). GSC-16021-1

Remote Data Exploration with the Interactive Data Language (IDL) Goddard Space Flight Center, Greenbelt, Maryland

ically in relatively complex algorithms(involving multiple threads), and thusshows the potential to relieve more ofthe programmer’s cognitive load as theproblem grows in complexity. Sequen-

ceL’s runtime environment then selectswhich parallelisms to actually exploit,with the aim of maximum overall speedwhen considering communication costsbetween processes.

This work was done by Daniel Cooke and J.Nelson Rushton of Texas Tech University forGoddard Space Flight Center. Further informa-tion is contained in a TSP (see page 1). GSC-15859-1

Mapping localized spectral features inlarge images demands sensitive and ro-bust detection algorithms. Two aspects oflarge images that can harm matched-fil-ter detection performance are addressedsimultaneously. First, multimodal back-grounds may thwart the typical Gaussianmodel. Second, outlier features can trig-ger false detections from large projec-tions onto the target vector.Two state-of-the-art approaches are

combined that independently addressoutlier false positives and multimodalbackgrounds. The background cluster-ing of Funk et al. models multimodalbackgrounds, and the mixture tunedmatched filter (MT-MF) of Boardman

et al. addresses outliers. Combiningthe two methods captures significantadditional performance benefits. Theresulting mixture tuned cluttermatched filter (MT-CMF) shows effec-tive performance on simulated and air-borne datasets.The classical MNF transform was ap-

plied, followed by k-means clustering.Then, each cluster’s mean, covariance,and the corresponding eigenvalues wereestimated. This yields a cluster-specificmatched filter estimate as well as a clus-ter-specific feasibility score to flag outlierfalse positives.The technology described is a proof

of concept that may be employed in fu-

ture target detection and mapping ap-plications for remote imaging spec-trometers. It is of most direct relevanceto JPL proposals for airborne and or-bital hyperspectral instruments. Appli-cations include subpixel target detec-tion in hyperspectral scenes formilitary surveillance. Earth science ap-plications include mineralogical map-ping, species discrimination for ecosys-tem health monitoring, and land useclassification.This work was done by David R. Thomp-

son, Lukas Mandrake, and Robert O. Greenof Caltech for NASA’s Jet Propulsion Labora-tory. For more information, contact [email protected]. NPO-48663

Mixture-Tuned, Clutter Matched Filter for Remote Detection ofSubpixel Spectral SignalsNASA’s Jet Propulsion Laboratory, Pasadena, California


Physical Sciences

The proposed quantum receiver inthis innovation partitions each binary sig-nal interval into two unequal segments: ashort “pre-measurement” segment in thebeginning of the symbol interval used tomake an initial guess with better proba-bility than 50/50 guessing, and a muchlonger segment used to make the high-sensitivity signal detection via field-can-cellation and photon-counting detec-tion. It was found that by assigning aslittle as 10% of the total signal energy tothe pre-measurement segment, the ini-tial 50/50 guess can be improved toabout 70/30, using the best availablemeasurements such as classical coherentor “optimized Kennedy” detection. However, 70% detection probability

(or, equivalently, 30% error probability)is not good enough for communica-tions, even with the most powerfulcodes, which require error probabilitiesin the 0.01–0.1 range to achieve the de-sired coded performance. Due to the re-quirement to maintain a constant-enve-lope local laser field, the recently

reported “optimized Kennedy” measure-ment was selected for making this initialguess. The outcome of this first measure-ment is used to decide which signal thereceiver should try to null. Hence, thelocal field envelope and phase are ad-justed to nearly cancel the more likelysignal, and photon counting is used forthe rest of the interval to confirm thisinitial decision. If the wrong signal is se-lected initially, then the local laser addsinstead of subtracting a constantmatched laser field to the received sig-nal, yielding a higher probability error;that is, higher probability of erroneouslypre-selecting the “other” binary signal.Optimum partitioning of the signal in-terval is critical, and must be carried outfor each new value of Ks. This conceptcan be extended directly to more thantwo intervals, by partitioning the first in-terval itself into two segments, optimizedfor the smaller initial energy, to furtherimprove the “pre-measurement” uponwhich the final high-sensitivity measure-ment strategy is based.

The performance gain of the parti-tioned-interval quantum receiver overthe well-known Kennedy receiver detec-tion strategy is shown in Figure 2, alongwith the gains of the classical coherentreceiver and the optimized Kennedy re-ceiver. It is noted that the coherent re-ceiver peaks at an average received pho-ton-count of Ks = 0.095 attaining amaximum gain of 1.272 over theKennedy receiver, whereas the opti-mized Kennedy receiver peaks at Ks =0.165, with a maximum gain of 1.381,after which both gains decrease as theaverage signal energy increases: the opti-mized Kennedy receiver approaches 1 athigh signal energies, reverting back tothe conventional Kennedy receiver,whereas the coherent receiver continuestowards zero. However, the partitioned-interval receiver described here attainshigher gains, and tends to maintainthese gains near their maximum valueeven with increasing signal energy. The only receiver structure known to

achieve the quantum limit theoretically

Partitioned-Interval Quantum Optical Communications ReceiverThe receiver structure described here improves upon the performance of optical communicationslinks, achieving high sensitivity and high data rates even with extremely weak optical signals.NASA’s Jet Propulsion Laboratory, Pasadena, California

Figure 1. Error Probability Performance and comparison of N-segment par-titioned receivers.

Figure 2. Gain of Coherent, Optimized Kennedy and partitioned receiversover Kennedy receiver.

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.810

-2

10-1

100

Classical coherentdetection

Kennedy receiver

Helstrom bound

Optimized Kennedyreceiver (N = 1)

Partitioned receiverN = 2

N = 3N = 4

Partitioned receiver performance:dB from Helstrom bound @ P(E)=0.1

N dB

1 1.122 0.783 0.484 0.41

Pro

bab

ility

of

erro

r, P(

E)

Average number of signal photons, 2=| α |sK

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.80.5

1

1.5

2

revi

ecer y

de

nn

eK r

evo

nia

G

Classical coherent detection

Optimized Kennedy receiver (N = 1)

Partitioned receiverN = 2 N = 3 N = 4

Helstrom bound

Pre

-mea

sure

men

t re

gio

n

Detection region

Average number of signal photons, 2=| α |sK


on binary signal detection (the curves la-beled “Helstrom bound” in Figures 1and 2) is known as the Dolinar receiver.This approach applies a rapidly time-varying local laser field to the signal dur-ing each bit-interval, but such time-vary-ing fields are difficult to generate inpractice at high data rates. In addition,the phase and sign of the local laserfields must be switched instantaneouslywith the detection of each new photonfor best performance, placing signifi-cant burdens on the processing speed ofthe receiver and on the response of thelocal laser. The proposed solution over-

comes these problems by employingconstant local laser intensities that canbe pre-computed based on estimates ofsignal-strength, while attaining nearlythe same bit-error rate as the more com-plex quantum-optimum receiver. Thesolution proposed here will thereforeenable high-sensitivity deep-space opti-cal co m muni ca tions at data rates up togigabits/second as required for futuredeep-space optical communications. This work was done by Victor A. Vilnrotter

of Caltech for NASA’s Jet Propulsion Labora-tory. For more information, contact [email protected].

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 321-1234800 Oak Grove DrivePasadena, CA 91109-8099E-mail: [email protected] to NPO-48496, volume and number


A multiple-pass optical platform elim-inates essentially all optical alignmentdegrees of freedom, save one. A four-pass absorption spectrometer architec-ture is made rigid by firmly mounting di-electric-coated mirror prisms with noalignment capability to the platform.The laser diode beam is collimated by asmall, custom-developed lens, which hasonly a rotational degree of freedomalong the standard optical “z” axis. Thisdegree is itself eliminated by adhesiveafter laser collimation. Only one degreeof freedom is preserved by allowing thelaser diode chip and mount subassemblyto move relative to the collimating lens

by using over-sized mounting holes. Thisallows full 360° motion of a few millime-ters relative to the lens, which, due tothe high numerical aperture of the lens,provides wide directional steering of thecollimated laser beam. Because the optical layout has been

designed to provide proper mirror align-ment for an orthonormal, paraxial laserbeam, this degree of freedom is suffi-cient to insure perfect optical alignmentonce the orthonormal condition is satis-fied. Further, the degree of freedom isenabled by using either simple loosemetal screws in the over-sized lasermounting holes, plastic screws with low

tension, or a combination of the two.Once alignment is achieved, the screwsare tightened sufficiently to insureruggedness, or the plastic screws may bereplaced, one by one, with metal screws.In either case, even the remaining de-gree of freedom is locked down after thefinal alignment. The final degree of free-dom may be permanently, or quasi-per-manently, locked by use of various adhe-sives on the screw head or threads.This work was done by Jeffrey Pilgrim and

Paula Gonzales of Vista Photonics, Inc. for God-dard Space Flight Center. Further informationis contained in a TSP (see page 1). GSC-16536-1

Practical UAV Optical Sensor Bench With Minimal AdjustabilityGoddard Space Flight Center, Greenbelt, Maryland

National Aeronautics andSpace Administration

technology focus electronics/computers - ntrs.nasa.gov · vectorized rebinning algorithm for fast...

Documents