proceedings annual advanced polymer dtic · advanced polymer components research symposium schedule...

PL-TR-92-3018 PL-TR--

AD-A25 6 679 92-3018

PROCEEDINGS OF THE FIRST ANNUAL ADVANCEDPOLYMER COMPONENTS SYMPOSIUM

John J. Rusek DTICS ELECTE

JUL 3 11992

A LPHILLIPS LABORATORYRKCPEDWARDS AFB, CALIFORNIA CA 93523-5000

July 1992

Interim Report

. .. .. .... . ... ....

92 7 Ip-, ()L- 53E

± PHILLIPS LABORATORYPropulsion Directorate

EDWARDS AIR FORCE BASE CA 93523-5000

NOTICE

When U.S. Government drawings, specifications, or other data are used for any purpose oth-er than a definitely related Government procurement operation, the fact that the Govern-ment may have formulated, furnished, or in any way supplied the said drawings, specifica-tions, or other data, is not to be regarded by implication or otherwise, or in any way licensingthe holder or any other person or corporation, or conveying any rights or permission to man-ufacture, use or sell any patented invention that may be related thereto.

FOREWORD

This Interim Report was submitted on completion of the First Annual Advanced PolymerComponents Symposium, Febuary 20-22, 1992, by the OLAC -PLIRKCP Branch at thePhillips Laboratory (AFSC), Edwards AFB, CA 93523-50. Project Manager for OLACPhillips Laboratory was Dr John J. Rusek

This report has been reviewed and is approved for release and distribution in accor-dance with the distribution statement on the cover and on the SF Form 298.

JOflN J. R6SEK BERGE GOSHGARIANProject Manager Chief, Propellants Branch

WAYN L. PRITZ RANNEY . ADAMSDirector Public A 1s Director(omponents Engineering Division

Acceoi For

NTIS CRA&ID-IIC 1A13U .ia:I;iot); ced JJustificdtion.........

By .. ......

Dist, ibution• I

Ava,',~IK ,

Dist

-*1I

-AI

Form Approv'edREPORT DOCUMENTATION PAGE OMB No. 0704-0188

,*.~,,~D:2 e~ -- ~::- I- a ~'a~ cx, oit espi encl~cing wh e u eior re% eA~ng n!-.ctions. searc-.'g eys 'nq data sources.-. -~ -: --~~ c-. .~':a~,,,,ee^-c ý! 2Iemono0' 'ýIrl'auon, Send ,mrnmenBs r,eo se' b 0er est~rnae or an, :se, et!Vecl lS

,- - r -, • - ',"" C.5 Ow' .a, n. "oscr -g aocr .ar.e,s Ser.,ces, *.rreC, orae ;o- 0'e n 'as r O r$e.Oc I'do CeDo' s, 1g l 5 JeftersonI_ C.' ;e ' •. •- --" -,. -', .- ,3:.. . ,: - . -- &:;. a'o f.• s:-c C'le " .'.-. c ert.e-, ant. £..o•eS. •'pf~~, •e•, c- ':Ct:0 ESC4•, ). W~aser¢cs.; . DCt Z•5C3.

1. A-2ENCs'&f C-1Y (!ee~e :!'2r'-) 12. PEPOrT DATE f3. REPORT TYPE AND DATES COVEREDI JULY 1992 Interim 21 April 19924. T!1E -:IND S.B'TITLE 5. FUNDING NUMBERS

Proceedings of The First Annual Advanced PolymerComponents Symposium PE: 62302F

PR: 5730

6. AL;7H-OR(S) TA: 00R9

DR. John Joseph Rusek

7. PERFCRM.,NG ORGANIZATION NAME(S) AND ADDRESS(ES) 6. PERFORMING ORGANIZATIONREPORT NUMBER

Phillips LaboratoryPL/RKCP PL-TR-92-3018EDWARDS AFB, CA 93523-5000

9. SPONSORING /MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSORING /MONITORING

AGENCY REPORT NUMBER

Phillips LaboratoryOLAC- FL/RKCPEdwards AFB, CA 93523-5000 PL-TR-92-3018

11. SUPPLEMENTARY NOTES

COSATI CODES: 21/06; 21/08/01; 21/09/01; 21/08/02; 21/09/02; 07/05; 07/0612a. DISTRIBUTION /AVAILABILITY STATEMENT ''12b. DISTRIBUTION C•ODE

"Approved for Public Release; Distribution is Unlimited"

13. ABSTRACT (Maximum 200 words)

Advanced propulsion concepts rely on advanced propulsion materials. The PhillipsLaboratory is aggressively pursuing advanced polymeric materials for use in Solid,Liquid, and Nuclear propul~ion component applications. Traditional composite mater-ials have high specific strengths, but suffer from high cost and labor intensiveprocessing. The APC program is currently exploring thermotropic liquid crystalpolymers. These materials have high specific strength and can be economicallyprocessed by traditional high volume routes such as injection molding and blow moldinApplications envisioned for these materials include rocket noZzles, pressure cases,propellant tanks and conduits, nuclear propulsion containmenta fairings, high pressurtanks and orbit-processed habitats for interplanetary voyages. These proceedingsdeal with an overview of liquid crystal polymer technology and current art, testarticle demonstrations and in-depth discussions of the so-called annealing phenomena.

.4. S Z,-_CT TER1.S '5. N.JJEER OF PAGESLiquid Crystal Polymers Liquid Propulsion Synchrotron RadiationAdvanced Materials Solid Propulsion Neutron Scattering |6 r--• CODEThermotropic Polyester Nuclear Propulsion Molecular Mechanics | ..

- .7y SECURITY CLASS;FICATION 19. SECURITY CLkS1I ,ICA ' . ''.' :nv OF AISTRACT"UnclOF THIS PAGEd OF AlSTRACT

Unclassified Unclassified Unclassified SAR' "' -' •" i : •-_-: 3 '•'• - -÷ : . . • - .' •_7

CONTENTSPREFACE iLIST OF PRESENTERS iiSYMPOSIUM SCHEDULE iiiEXECUTIVE SUMMARY 1PROGRAM TASKING 3GLOSSARY 7

APC OVERVIEW, J. Rusek 8

POLYMER SYNTHESIS, J. Rusek 20

CRYOGEN TESTING, K. Chaffee 31

MOLECULAR MECHANICS, S. Lieb 54

ATOMIC FORCE MICROSCOPY, D. Silver 66

SURFACE SPECTROSCOPY, J. Mann 74

SYNCHROTRON STUDIES, R. Hoffman 90

REFLECTANCE SPECTROSCOPY, P. Oldham 113

MICROMECHANICS, A. Palazotto 140

DIELECTRIC SPECTROSCOPY, D. Kranbuehl 182

RHEOLOGY, D. Schwartz 207

MATERIALS DATABASE, T. Elkins 217

MACROSCOPIC ANALYSIS, J. Shelley 256

THERMAL ANALYSIS, P. Jones 269

X-RAY/NEUTRON REDUCTION, S. Osborn 286

MOLD DESIGN, T. Elkins 301

PART DESIGN, C. Frank 314

PROCESS RESEARCH, N. Schott 325

LOW TEMP CVD/COMPATABILITY, E. Wucherer 326

2X4 MOTOR, H. Nguyen 343

LASER EARTH TO ORBIT VESSEL, J. Kare 355

FLIGHT LOGISTICS APPLICATIONS, S. Hardy 368

RETROFIT APPLICATIONS, T. Fuchser 370

RETICULATED BONDLINE, K. Mahaffy 371

PREFACE

The Advanced Polymer Components Initiative began in December of1989. The initial purpose of the program was to explore advancedengineering polymers for use in propulsive applications. Threemain objectives were established: apply commercially availablematerials to solid propulsion needs, apply these materials toliquid propulsion needs and establish a design capability for thenew processing techniques envisioned to produce parts. Thermo-tropic liquid crystal polymers such as VECTRA, XYDAR and otherswere identified as having the most promise for surviving therigors identified with space propulsion applications. Within thefirst few months of the program, it became apparent that thesenew materials needed significantly more research before finalpart fabrication could be accomplished.

A team of over two dozen researchers was assembled to attackthe fundamental questions raised. The researchers were chosenfrom government, academia and industry, both nationally andinternationally, to focus on specific research areas to quicklyenable the Air Force to use these novel materials.

After two years of research, the First Annual AdvancedPolymer Components Symposium was held. This initial meeting wasconducted at Butler University in Indianapolis, Indiana fromFebruary 20-22, 1992. Twenty-four researchers and Air Force userswere in attendance from accross the country. This document is theresult of that meeting.

In addition to the executive summary and task plan, thisreport contains the abstracts and content of the twenty-tworesearch papers and Air Force user input. The intent of thereport is to indicate the status of Air Force propulsion researchusing thermotropic liquid crystal polymers. It is anticipatedthat a second symposium will be hela the spring of next year.

The United States Air Force is indebted to all of the par-ticipants on this critical program for their goodwill and dedica-tion to this effort.

John J. RusekProgram ManagerAdvanced Polymer ComponentsInitiative

APC SYMPOSIUMPRESENTER LIST

Ms Rebecca At Mr Poul Jones Prof Phil OldhamMaterials Science Department OLAC PL./RKFC Chemistry DepartmentNorthwesteorn University Edwards AFI CA 93523 Mississippi State UniversityEvanston IL 60206 T(M0)275-5315 Mississippi State MS 39762T(708)491 -3996 FC805)275-5144 T(601)325-3584FC708)491 -7820 F(601 )325-7807

Dr Jordin KareOr Kevin P. Chaffee L-278 PO Box 808 Mr Stephen OsbornOLAC PL/RKCP 1.1111 OLAC PL/RKCPEdwards AFI CA 93523 Livermore CA 94550 Edwards AFB CA 93523T(805)275-5740 1(510)423-8300 T(805)275-5741FC805)275-51"4 F(510)423-9178 F(805)275-5144

Mr Tem A. Elkins Prof David Kranbushl Prof Tony PalazottoOLAC PLIRKAD Chemistry Department Astronautics DepartmentEdwards AFI CA 93523 College of William and Mary Air Force Institute of Technotogy1(805)275-5303 Williamsburg VA 23185 Wright-Patterson AFB OH 45433F(805)275-5144 TC804)221-2542 T(513)255-2998

FC804)221-1021 F(513)476-4055Mr L. Jackson EtheridgeOrganic Technologies Prof Shannon Lieb Or John J. RusekColumbus, 0ON 43220 Chemistry Department OLAC PL/RKCPT(614)459-5000 Butler University Edwards AFB CA 93523F(614)459-4372 Indianapolis IN 46208 T(805)275-5407

T(317)283-9410 F(805)275-5144Mr Chris Frank F(317)283-9519SM AL/TIEC Mr Daniel F. SchwartzMcClellan AFS CA 95652 Mr Kevin E. Mahaffy OLAC PL/RKCPT(916)643-3810 OLAC PL/RKCC Edwards AFB, CA 93523FC916)U3-680 Edwards AFS CA 93523 1(805) 275-5183

T(805)275-5629 F(805) 275-5144Mr Terry Fuchser F(805)275-5144Mercer University Ms J. ShelleyEngineering Research Canter Prof J. Mamn OLAC PL/RKCCWarner Robbins, GA 3109 Chemical Engineering Department Edwards AFB CA 93523T(912) 929-M00 Case Western Reserve University T(805)275-5394'F(912)929-6406 Cleveland ON 44106 F(805)275-5144

T(216)368-4150Cpt Steve Hardy F(216)368-3016 ILt Dave SilverQO-ALC/TIELM OLAC PL/RKFCHill AFB UT 84056 Prof Laurence Marks Edwards AFB CA 93523T(801)777-2874 Materials science Department T(805)275-5410F(801)777-8049 Northwestern University F(805)275-S144

Evanston IL 60208T(708)491-3996 Dr E.J. Wucherer

Prof Dick Hoffman F(708)491-7820 OLAC PL/RKFCPhysics Department Edwards AFB CA 93523Case, Western Reserve University Mr Mieu Nguyen 1(805)275-5759Cleveland ON "4106 OLAC PL./RKCC F(80)275-51441(216)369-4012 Edwards AFB CA 93523F(216)368-4671 T(805)275-5629

F(8OS)275-5144

ADVANCED POLYMER COMPONENTS RESEARCH SYMPOSIUM SCHEDULE

O 19 FEB1800-2000 Check In

20FEB0800-0830 APC Overview- J. Rusek0830-0900 Polymer Synthesis- J. Rusek0900-0930 Polymer Analysis- K. Chaffee0930-1000 Molecular Mechanics- S. Lieb1000-1030 Break1030-1100 Atomic Force Microscopy- D. Silver1100-1130 Surface Spectroscopy- J. Mann1130-1200 Synchrotron Studies- R. Hoffman1200-1300 Lunch1300-1330 Reflectance Spectroscopy- P. Oldham1330-1400 Micromechanics- A. Palazotto1400-1430 Dielectric Spectroscopy- D. Kranbuehl1430-1500 Rheology- D. Schwartz1500-1530 Materials Database- T. Elkins1530-1600 Break1600-1630 Macroscopic Analysis- J. Shelley1630-1700 Thermal Analysis- P. Jones1700-1730 X-ray/Neutron Reduction- S. Osborn1730-1800 TEM of Advanced Polymers- L. Marks

O 21FEB0800-0830 Mold Design- T. Elkins0830-0900 Part Design- C. Frank0900-0930 Injection Molding- R. Griffin0930-1000 Process Research- N. Schott1000-1030 Break1030-1100 Low Temp CVD- E. Wucherer1100-1130 2x4 Motor- H. Nguyen1130-1200 Subscale Flight Demonstrator- A. Kenny1200-1230 Laser Earth to Orbit Vessel- J. Kare1230-1330 Lunch1330-1800 Polymer Characterization Discussion

22FEB0800-1200 Polymer Processing Discussion1200-1300 Lunch1300-1800 Polymer Applications/Future Research

III

EXECUTIVE SUMMARY

The advent of advanced composite materials based on graphite andKEVLAR fibers has shown production articles with very high spe-cific strength and modulus values. The main drawback to thesearticles are their cost and labor intensive processing. Thermalprocessing of plastics yields inexpensive parts, but at a signif-icant lowering of performance, in most cases. Thermotropic liquidcrystal polymers exhibit ordered behavior in the liquid state;this order can be frozen into the structure of the part thusyielding a "molecular composite" where processing becomes thedeterminant of the ultimate part strength.

The Advanced Polymer Components Initiative was begun overtwo years ago to study these advanced materials. Initially,commercial resins were injection molded into test articles forevaluation at the Phillips Laboratory. One anomaly was noted assoon as articles were produced; the properties of the skin of thepart were far different from the core region. Sectioned specimenswere analyzed and found to be different not only in strength, butchemical and thermal properties as well;most polymers exhibitthis, but not to this degree. Chemical compatability with mono-methylhydrazine is enhanced dramatically in the skin region.

A second observation is the so-called annealing phenomenon.A part derived from certain resins can be heat treated below themelt transition to yield enhanced solvent resistance and adramatic increase in the melt temperature. This phenomena couldpotentially yield rocket nozzles with low erosion, propellanttanks and conduits for use in liquid rockets and moderators andworking fluid containment for advanced propulsion scenarios.

On 20-23 February 1992, Phillips Laboratory researchersattended the first national meeting in support of the AdvancedPolymer Components Initiative. The purpose of the meeting was toprovide an open forum where all researchers and potential userscould address issues concerning the use of thermotropic liquidcrystal polymers as structural materials for Air Force propulsionapplications. The main topics of discussion were fundamentalproperties, material processing (part design and fabrication),applications and planning.

The symposium was attended by twenty-three engineers andscientists representing three Air Force installations, sevenuniversities, two corporations and one national laboratory. Theresults or progress reports for each APC task were presented bythe task managers. The discussions were logically directed to-wards understanding how the raw materials fundamental propertiesaffect processability and how the processing techniques in turndetermine the mechanical, thermal and chemical properties of thefinished product.

The rheological and thezmal degradation characteristics ofthe commercially available polymers are sufficiently understood

to permit test sample production by injection molding. However,the synthesis procedures employed by the manufacturers appear toyield polymer resins with variable impurity content. The polydis-persity and the block distribution of the copolymers are alsounknown. It is thus difficult to adjust the injection moldingparameters to optimize the part properties in a consistant fash-!on. This fact was made quite evident by the scatter of themeasured mechanical properties of injection molded test speci-mens. One area decided to be of key import is the synthesis ofwell characterized model polymers.

Like most polymeric materials, the properties of thesemolecular composites depend upon the specific mechanical andthermal history. Where this is most evident is in the creation ofthe annealed polymer. The formation and structure of the annealedregions are not well understood. This phenomena is the key topotential applicability of these materials as structural materi-als and therefore the area where most of the fundamental researchin the future will be directed.

Comments elicited after the symposium from all attendeeswere highly positive. The brainstorming sessions yielded a pleth-ora of new approaches. It is anticipated that a second symposiumwill be held in the spring of 1993 to continue the communicationon this critical research effort.

2o

ADVANCED POLYMER COMPONENTS PROGRAM TASKING

FUNDAMENTAL CONTRACT TASKS

TASK #1 MOLECULAR MECHANICSS.LIEB, Butler University

Analyze intramolecular minimum energy configurations formodel rigid rod polymers. Resolve intermolecular forces thoughtto occur during polymer annealing.

TASK #2 SURFACE SPECTROSCOPYJ. MANN, Case Western Reserve University

Analyze polymer surfaces using ellipsometric FTIR and Ramanspectroscopy to deduce orientation effects in the annealingprocess. Analyze surface structure of annealed polymers by x-rayreflectivity.

TASK #3 REFLECTANCE SPECTROSCOPYP. OLDHAM, Mississippi State University

Analyze polymer/solution interfaces by means of the totalinternal reflection fluorescence phenomena. Understand solvolysisof ordered polymers.

TASK #4 MICROMECHANICSA. PALAZOTTO, Air Force Institute of Technology

Obtain mechanical property information in the gauge betweenmolecular and macroscopic size domains. Correlate microscopicphenomena to molecular theory and macroscopic results obtained inallied tasks.

TASK #5 DIELECTRIC SPECTROSCOPYD. KRANBUEHL, College of William and Mary

Determine the macroscopic states, phase transition tempera-tures and molecular/morphological structure of liquid crystalpolymers using dielectric spectroscopy and thermal analysistechniques.

TASK #6 PROCESS RESEARCHN. SCHOTT, University of Lowell

Obtain commercial resins, mold test specimens as a functionof process parameters and perform mechanical testing to directlycorrelate macroscopic properties to polymer rheology. Explorealternate processing techniques.

FUNDAMENTAL IN-HOUSE TASKS

TASK #7 POLYMER SYNTHESISJ. RUSEK

K. CHAFFEESynthesize model liquid crystal polymers with varying pend-

ant groups in support of the annealing research tasks. Character-ize these compounds by LALLS, FTIR, NMR and VPO.

3

TASK #8 POLYMER X-RAY ANALYSISJ. RUSEK

K. CHAFFEES. OSBORNAnalyze near surface annealing phenomena via low angle x-ray

diffraction, high-Z substituent EXAFS and kinetic studies byQEXAFS in full collaboration with researchers at DESY.

TASK #9 POLYMER NEUTRON ANALYSISJ. RUSEK

K. CHAFFEES. OSBORNExplore the annealed state as a function of depth on as-

processed and deuterated polymer samples; study the kinetics ofthe annealing phenomena in the liquid state in full collaborationwith researchers at ANSTO.

TASK #10 ATOMIC FORCE MICROSCOPYD. SILVER

Observe equilibrium polymer configuration at the atomiclevel. Correlate polymer morphology with annealing phenomena andsupport the polymer processing studies.

TASK #11 CHEMICAL VAPOR DEPOSITIONE. WUCHERER

Determine the feasability of low temperature CVD pLocessesto coat advanced polymers with aluminuum, nickel and 'otentiallyrhenium. Understand mechanisms of adhesion as related to the CVDprocess.

TASK 112 RHEOLOGYD. SCHWARTZ

Ascertain appropriate processing conditions for injectionmolding of liquid crystal polymers by means of mechanical as wellas dielectric spectroscopy. Observe and predict cooling anomaliesgermane to the annealing phenomena.

TASK #13 MACROSCOPIC ANALYSISJ. SHELLEY

P. JONESC. FRANK, McClellan AFBDetermine the engineering properties of liquid crystal

polymers by mechanical and thermal testing. Assess the sensitivi-ty of these polymers to design variables and gauge their strengthefficiencies.

TASK #14 MATERIALS DATABASET. ELKINS

Construct a dynamic data storage and retrieval system formechanical, chemical, thermal, electrical and environmentalproperties. Specific emphasis is placed on neat resin properties,both as-molded and annealed.

4

APPLICATION AREA TASKS

TASK #15 INJECTION MOLDINGC. FRANK, McClellan AFB

S. HARDY, Hill AFBR. GRIFFIN, Hill AFBDesign and fabricate tooling to produce injection molds in

support of all phases of the APC program. Injection-mold testarticles and final flight products for all allied tasks.

TASK #16 STATIC 2X4 DEMONSTRATORH. NGUYEN

C. FRANK, McClellan AFBR. GRIFFIN, Hill AFBStudy the pressure and ablation effects of liquid crystal

polymers by means of the standardized 2X4 test Lixture. Quantifythe heat transfer properties of the polymers by precise thermalmeasurements through the case. Results will point to useage ofthe LCP's as solid rocket motor cases and large launch igniters.

TASK #17 SUBSCALE FLIGHT DEMONSTRATORH. NGUYEN

C. FRANK, McClellan AFBR. GRIFFIN, Hill AFBDesign and fabricate nominal 100 # thrust boosters incorpo-

rating total polymer case and nozzle closures. Nozzle assembliesare CVDed to reduce erosion; boosters will be launched at Colora-do Springs CO. Results will assess flight-weight useage.

TASK #18 BLOW-MOLDINGJ. SHELLEY

J. RUSEKC. FRANK, McClellan AFBH. COX, General Motors ResearchI. ABU-ISA, General Motors ResearchAssess the feasability of thermotropic LCP's as blow-molding

resins to impart biaxial orientation to cast pressure vessels.Test these vessels to failure with defined liquid propellants.Results will show potential as OTV high pressure tanks, Marsmission shrouds and LLV propellant tanks.

TASK #19 LASER EARTH TO ORBIT DEMONSTRATORJ. KARE, Lawrence Livermore National Laboratories

J. RUSEKJ. SHELLEYC. FRANK, McClellan AFBDesign and produce a biaxially oriented 2-liter hydrogen

storage vessel using blow-molding technology and commercialLCP's. Launch test vehicle at Kirtland AFB using MW-class chemi-cal lasers. Results will point to useage in the storage ofliquid hydrogen and tethered flight operations.

5

TASK #20 NUCLEAR PROPULSIONJ. SHELLEY

J. RUSEKK. CHAFFEEC. FRANK, McClellan AFBAssess the feasability of LCP's as candidate materials for

nuclear propulsion. Thermal properties will be measured usinglaser thermal analysis techniques. Results will point to contain-ment for nuclear propulsion applications.

0

GLOSSARYAFM-Atomic Force MicroscopyANSTO-Australian Nuclear Science and Technology OrganisationAP-Ammonium PerchlorateAPC-Advanced Polymer ComponentsASTM-American Society for Testing of MaterialsCHQ-ChlorohydroquinoneCVD-Chemical Vapor DepositionDBMS-Database Management SystemDESY-Deutsches Elektronen-SynchrotronDMA-Dynamic Mechanical AnalyzerDSC-Differential Scanning CalorimetryEDC-Ethylene DichlorideEXAFS-Extended X-ray Absorption Fine StructureFDEMS-Frequency Dependent Electromagnetic SensingFEA-Finite Element AnalysisFTIR-Fourier Transformed Infrared SpectroscopyIDEAS-Integrated Design Engir.3ering Analysis SoftwareIR-InfraredLALLS-Low Angle Laser Light ScatteringLB-Langmuir-Blodgett FilmLCP-Liquid Crystal PolymerLH2-Liquid HydrogenLN2-Liquid NitrogenLTA-Laser Thermal AnalyzerMe-MethylMHQ-MethylhydroquinoneMMH-MonomethylhydrazineNEMESIS-New Eng. Materials Eval./Surface & Interface StudiesNMR-Nuclear Magnetic Resonance SpectroscopyNTO-Nitrogen TetroxideODE-ordinary Differential EquationsOTV-Orbit Transfer VehiclePAS-Photoacoustic SpectroscopyPBO-PolybenzoxamidePBZT-PolybenzthiazolePEHQ-2(1-Phenylethyl)hydroquinonePh-PhenylPhEt-PhenylethylPHQ-PhenylhydroquinonePMT-Photomultiplier TubeQEXAFS-Quick Acquisition EXAFSRDA-Rheometrics Dynamic AnalyzerRF-Radio FrequencySDRC-Structural Dynamics Research CorporationTA-Terephthalic AcidTEM-Transmission Electron MicroscopyTIRF-Total Internal Reflection FluorescenceTMA-Thermal Mechanical AnalyzerUV-UltravioletVATIRF-Variable Angle TIRF AnalysisVPO-Vapor Phase OsmometryXAFS-X-ray Absorption Fine StructureXANES-X-ray Absorption Near Edge StructureXRD-X-ray Diffraction

7

APC OVERVIEW

J.J. RusekOLAC PL/RKCP

Edwards AFB CA 93523

Advanced propulsion concepts rely on advanced propulsionmaterials. The Phillips Laboratory is aggressively pursuingadvanced polymeric materials for use in solid, liquid and nuclearpropulsion component applications. Traditional composite materi-als have high specific stengths, but suffer from high cost andlabor intensive processing. The APC program is currently explor-ing thermotropic liquid crystal polymers. These materials havehigh specific strength and can be economically processed bytraditional high volume routes such as injection molding and blowmolding. Applications envisioned for these materials includerocket nozzles, pressure cases, propellant tanks and conduits,nuclear propulsion containments, fairings, high pressure tanksand orbit-processed habitats for interplanetary voyages.

This paper will deal with an overview of liquid crystalpolymer technology and current art, test article demonstrationand an introduction to the so-called annealing phenomena. Final-ly, next-generation space applications will be discussed germaneto the Space Exploration Initiative.

8

I

C!

C/ /

cc 2

I0 wL i l

LJcn

zz LLi0W0 CL 0

Co L0 0mem- - w cc jn

I- D - .- r- UZjLL~w w Ci

010gU U.

0 C/) 9

(I)

LL LU-j CL m ..

10~~I

LLLJLco

LIJ 1 I9- C- -- _

o -JJ LI.

a.H

00 UU

X U .- I<0

00

0w

00

z0 ICl)wL N 0 Owz ý

0 13

* 0wC

00

( 4. L I

ui~

0

C11

w0

o zC/)

zz0) wzz00

coz w - ý,

I->oW 0 Z

0 L. W

0 -J1

0 F',)

0 cr 0. =0 0)2 o uU

W. ) _j -I4

S0 0 ý MO d u.J

4 0 x <- =CL Z., Z-

U U

41-

12

-4)00

C/) LU-

zL I--LULLzz

LLLU

0J~j U

0 U~z a zO< <OOHF Z L. u<

0-4 H-- <ULu:Dcr <U CL)

Z 0 LLUcr-

< ZWUU LLQ >

zL 0Q2D- u-j I

0.L LLUJF--

7Z- 13

4*)

z LUJzz <U

0 U-J

o~L LU --(in)J u a

u 0J)~H ~L

L H< F- cc U000U Ix 0L

0 z H L )- H r F- F-X F-LLU-11 Lu~u <U 0N0: ci) < U) 0 0 H i -

z) < -I LL U< <

z

LU

-J

H 0.Z~0~

V)U'

0 uu 00-

~L -1 0 00<- Lf <Ez c Lo LU) 000

m LUJz0 0 >ULU a 0. ~XLL

0 U0. a 0

<~ ULUl UU 0 zZ >U UZJD -> c X xLLi Uoo 0 D F- oF-~

=c Z >n = U - m UL ~ L ) U U<z Z ZU'

< -<UJ LLL.J L

LU<z Z Lo <H U' iD

>~~j CO 0 P o

0 < uoIc)ua u xL C

Lu F-M < _ Z15

CI

0

-z

I-o 0.z

= -A

dd uj

cr-j

U- .

Cw (n mwP.

cc Z 0L~

z w)

16V

z0

.... ...... 0

C/) .

E m

Ww

18

POLYMER SYNTHESIS

J.J. RusekOLAC PL/RKCP

Edwards AFB CA 93523

A concept coined polymer annealing was observed in DuPontHX-4000 and Granmont GRANLAR. These polymers, when appropriatelyheat treated, exhibit significant elevation of the melt tempera-ture and greatly increased solvent resistance. The HX-4000 couldbe thought of as chain extending during treatment, since it issynthesized from the melt, however, the GRANLAR species is fullyreacted in solution and exhibits an obviation of the melt entire-ly. It is thought that the size and the polarity of the pendantgroup in the polyester is the prime determinant in the ability toanneal.

This paper will address the annealing phenomena by lookingat the molecular structure of the liquid crystal polyesters.Synthetic schemes are presented as well as the polymers synthe-sized to date. The critical nature of polymer synthesis as itsupports the rest of the APC effort is stressed.

0

020

0

C)C) ______________

aa

0 NN� w0% N%% �ccC. C) Owz

0 0 �cI)w N fl Ow2 'U

0 m Wa,z�O an

0o0

w1,

-40Pa -

CI aI0 wo r0 1

2 1

0

L -i -

6*

i�9*

V.

3.'

23

B

t )�jgp

* 9 g 9 ei 24

U)

zz

LL UJ

Eu

a IL

IZWZAoj

~I~m CC

0 E

'J .;.N�

S

'mm

Ias

I

A�

VI

SI

I

'4

I

I

'I�.

C

I J

-

)0

F

H

00 -

N.

=

27

el&

00

28 A

vU -. N"'41 ~ P

*~,'tw

w wi

29

% IL

'aý'

.. .. .. .

CRYOGEN TESTING

K.P. ChaffeeOLAC PL/RKCP

Edwards AFB, CA 93523

ABSTRACT

The objective of tasks 12,13 and 14 was to examine themechanical integrity and stability of the commercially availablethermotropic liquid crystal polymers (LCP) at cryogenic tempera-tures. The liquid oxygen (LOX) testing vas conducted at the NASAWhite Sands Test Facility by Harold Beeson and Richard Shelley.The liquid hydrogen (LH2) mechanical testing was performed by TomEisenreich of General Dynamics Space Systems Division. Finally,the liquid nitrogen (LN2) burst tests were performed at thePhillips Laboratory by Eric Schmidt.

The LOX test results imply that the injection molded commer-cially available LCP materials are not suitable for liquid oxygenuse. Although the materials compared favorably to Teflon (PTFE)in the autoignition test, the LCPs relatively high reactivity tomechanical impact and combustible gaseous emissions make thempoor canidates for tank materials. Vectra A950 had the overallbest properties while DuPont HX4000 had the poorest.

The longitudinal XYDAR SRT-500 had the highest tensile andflexural strength and modulus at LH2 temperatures. All materialsdisplayed anisotropic mechanical behaviour, as expected. GeneralDynamics reports considerable scatter in all results. This isperhaps a result of the injection molding process.

VECTRA A950 had the highest ambient temperature burst pres-sure of 7.38 MPa while the XYDAR SRT-300 had the highest LN2burst pressure at 9.21 MPa.

S

(ccW --(.

IOLLL

0jQ-

32

CC/z Coz

Cl/) I

H--

0Z Z H-Co CO

x NJ C\j

33

0

C. EoE o-- o-

('5

o E

0 E , 1....

"0

34

C/)

wC/oz0

F- c) Q00

~~C:) 0

0.<--

I cr.0J 4". 0Oc o

W Hl

35

ELECTROKAGmET

II

PLUMMIET ASSEMBLY

TIMER PICKUP

STRIKER PIN

& GUIDE

CUP ASSEMBLY

SL CONCRETE BASE

Figure 1. Mechanical Impact Testing Apparatus

36

a

c E C:

0 0 41

+10E 0

p ~U

0.0

a 0r Ii ;

IL

0

L

LI I-- •I-

W 0; ; ---------

---- i J

oi L6> C~*

L L

V 0.

o

*

0

c N>0

37

N-oc

CL~)

nNE-zjm

C)'0V -

'a .- < 0*

0 0~

--W

0 0

00

00

38

Ignition Wire

S~Test Sample

Protective Di u

Test ChamberBase Plate

Figure 4. Promoted Combustion Test Apparatus

S39

u 0

m r

m

0U')

Ln

a)

0 0

000

U..

m cu

to

on rLi i) 0L oL

140 40

i .00

0ci 0

NN

0 0

cr4

m0

ro ICLo

to o

4-WI

4.'41

U,)

0-0U,)

S-%-

t Li

0L 0

M - Aj L-41 CL~

all3 -

03

co 0

m.44

o uuQ4 In 0 '

cuo. l

(3S139)82 vIStiIO

442

CCCC

u ,~ILi

Va40:C,

C I

..- a

-S a.-

5- I

1.1~ap-

CD co00a% co -0

(261).ainsss~c

PL04SDJ4

043

oCD 0

0x

0o

c'J

0

COD

00

o -

co

o ot0- S0

S.-,

CD u

0

84

CDC

o C7

o 1.wo -S

V!~~(% LI ! 0! L!c

44.

P0

z

w

.j-

CO)

LL

45

I* ie _ i ii elIC

f9t

Z+ c

aft.,.y •_-.

PI• < - III 3_

46

-i - - -

V0 . cx

qcl d~r

- -

a 47

14 Ilo It 4 IinX17-

48L

m

- a

-- a

I-*1�

.4 -

E�7 �..- -

�%rf ff.,..r... � -'S �

'I,

-S

'S

- -

* N

4'-S

��

0 -

.5'

�

I.. �* .Z

a � 5� -�

'-S

= -. 4'5

-a'J

-- 'I � :�****'.-.��5 �

'---�:'�*

- J.4�$ -� � * �

- ..-.- '* -*" '-5-

�4 L� �

0 LI 0 1 LIN N

11�r�i!�P�c �3�JJ�I -�

.5-.

49 c�i

5.-

Sa9 -S9

(jP P

C4 C- M- C4r .. a

~ %kOON'~ *%% "Z tc -

CA - ~

N Ng= -i en4 r* ca De 0,

;A 'n 'let~ ~V)

I .-T -.. ) on%_______________

I.- , I. %

C; c-. d -6 -;4 i c

o AR 4~4( M~ ~4re'~ ~ct

o - -- -c~ 5-

- 0

S=. ,- .• .- -- .,

0 2• d, C .,s

'fl

gg

~1~~ ~r-

0., =•

-~ c4 $V

oo09ibs Wvagoo09i~s wYag00O1S WYGAX

oo09i~s Wvag&:00918S 8vawOOM1S WagAOO0CS MYAXoo0UHS 8vwxoo0c1~s WYagoo0cis MYAXOOM1S wagA

:00ci~s 8vag

C14 :OLZ 0 MAX

:98OJI38AC :0968 0113Aj~

0968 0JI33AQ~II? 0g6a DJ138A

:09 O68 OJIMA

0952 DJ)OaAO096V 0JI33AO096V 0438AO096V 01138A

404 opolO:66 0~JI3A

-aO~V 0'f-8A

5223

LN2 Burst Test Reference:

"The Application of Liquid Crystal Polymers to Turbomachinery",M.A. Mueller and E.E. Schmidt, Proceedings of the 1992 JANNAFPropulsion Meeting, Indianapolis IN.

S

53

Presenter: Shannon G. LiebDepartment of ChemistryButler University4600 Sunset AvenueIndianapolis, IN 46208

Phone: (317) 283-9410FAX: (317) 283-9519Bitnet: Lieb@ButlerU

Abstract

Molecular Dynamics Investigation of Polymer Annealing

Polymers have an essential use in the making of casings ofsolid rocket boosters. An investigation into the molecular structureand the resulting macroscopic properties is necessary in order tomodify existing technology for the production of casing material forpresent and future uses. The "annealing" process of some liquidcrystal polymers (synthesized from a monosubstituted hydroquinoneand terephthalic acid) produces a polyester whose physicalcharacteristics do not seem attributable to the usual further "cross-linking" explanations. The annealing process of holding the polymerat a temperature just below the melting point for five or so hoursproduces a material that does not me!t at its original melting point,but can be raised to its decomposition temperature without melting.The label of order-disorder phase transition seems a moreappropriate term for this process. The object of this theoreticalinvestigation is to uncover the importance of the substituent of thehydroquinone in the action of producing this phenomenon. Thisongoing investigation is divided intn two tracks. The first is amolecular dynamics simulation of the annealing process. The secondis the use of of some spectroscopic techniques to aid in the properparametrization of the theoretical modeling, as well as, lending somebench marks by which the theory can be gauged. As the projectunfolds it is anticipated that the theoretical modeling will serve as apredictive tool in suggesting other liquid crystal polymer structuresthat may or may not have this phase transition inherently built intothem.

054

SFfi

m m n

Figure 1. A typical monomer unit for a monosubstituted hydroquinoneterephthalic acid polyester. The R group can be in either of the two uniquepositions available to it in the hydroquinone portion of the structure. Themodel systems proposed for study have R = CH2-CH2-C6H5 (phenethyl), CH3(methyl) and Cl (chloro).

As a liquid crystal polymer this material behaves as a rigid rodmolecule. Interestingly, of the three polymers shown in figure 1(differentiated by their "R" groups) the phenethyl derivativedemonstrates the annealing behavior and the other two show only a"partial annealing". Because of the rigid rod characteristics and lackof many functionalities (only the two end groups) left over for thetypical cross-linking, it seems more appropriate to view this as aphase transition (i.e., an order-disorder phase transition). It isfurther my contention that this phase transition is enhanced by thefreely rotating phenethyl group interacting with neighboringaromatic rings whether it be those found on the polymer backboneor other phenethyl groups. The crystal structure of benzene hasbeen found to have the benzene molecules stacking such that theedge of one ring is perpendicular to the face of its nearest neighborsas depicted in figure 2. 01Figure 2. The perpendicular arrangemen, of two benzene molecules as foundin the solid state. The second benzene molecule is to the right with its edgecoming directly out of the page.

Introduction

This problem is best modeled at a theoretical level using ageneral approach that has been explored for the past 15 years or soby those interested in biopolymers (i.e., biochemists). The techniqueused in that case was to acknowledge that the problem was well

55

beyond the computational limits of ab initio (first principles)calculations and one had to rethink what questions could beanswered with the computing power available. The obvious answeris to push the techniques that made molecular modeling"fashionable" to begin with. The elucidation of the double helixstructure of DNA was in no small part accomplished by the use of"hand-held" molecular models which piece together and indicate the3-dimensional ramifications of placing atoms together in moleculesusing accepted bonds lengths and angles for various combinations ofatoms. There are several obvious omissions from this approach.Although torsional motions (those motions accompanying rotations ofgroups stuck to either end of a bonded pair of atoms) are allowed,there is no restriction of motion about a single bond which is anunpleasant event for long stretches of singly bonded atoms; takepolymethylene for example. There is no convenient way torepresent interactions between different molecules. There is anaverage bond length used to represent every pair of atominteractions. Even though the average length takes into account thenature of the two atoms and the nature of the bonding between them(i.e., single, double or triple bonds), it does not allow the bond angles,lengths and torsionw to readjust to each new chemical environment.The next obvious step is to devise a computer image of the molecularstructure under question and set down rules for accommodating eachof the aforementioned drawbacks. This extension allows a very widelatitude for development of a molecular model. The obviousdrawback is that the model is Newtonian - that is, it creates amolecule that is a classical mechanics model of the molecule ratherthan a quantum mechanical model. One can, in part, set thatobjection aside by insisting on the use of inter and intramolecularforces that are based on quantum mechanical potential models thathave developed over the years. The issues that remain unsettled arethe violation of the Uncertainty Principle which does not allow forthe trajectory description of the classical model and when does aclassical model begin to mimic a statistical average of a largeensemble of molecules. These philosophical issues need to be born inmind while pushing of the frontiers of the relationship betweenstructure and reactivity - the keywords found at the front ofevery introductory general chemistry text.

So much for background and philosophical issues, thereremains the description of the Newtonian mechanics applied tomolecular systems. This is referred to as Molecular Mechanics. Thequantum mechanical potentials typically used are shown on the nextpage.

56

* Molecular Mechanics Potentials

Non-bonding interactions

a Ve~ (!) 12 ( 6!) + r (1-6412)

Bonding interactions

b I I~kb (b -bd 2

Vod2 1, 2SV IC.. = I I ke (e - 2o

Vbond 2 0 e-2Ou

Vtorion = Xk,[ l+cos(n*+8)]

Van der Waal radii are used to detect when atoms arecoming too close together and a high repulsive energy penalty isadded for interactions which infringe on the van der Waal sphere.

To cut down on the amount of computation, extendedatom approaches can be applied for C-H groups (i.e., a C-H or CH2 orCH3 group can be replaced by a single "extended" atom)

Newton's equations of motion for quantum mechanicalpotentials:

2

m. = - Vi[ U(rl,r 2 , ... ,rN) i1= INidt2

One solves the N coupled differential equations numerically to obtainthe trajectories of the atoms of the system.

The definition of the temperature of the system at some time, t, isdefined as:

1 N 2(3N-n)kB Im, dII

where kB is the Boltzmann constant, 3N-n is the number ofunconstrained degrees of freedom in the system and vi is thevelocity of atom i at time t.

I57

By way of explanation, one can see that the first equation onthe previous page uses the Lennard-Jones 6-12 potential to describe 0the dispersive forces between molecules and an electrostatic term isadded in as well because those forces are of equal or g:eaterimportance in describing intramolecular interactions. Since we arefree to make the charges (qi and qj) non-integer, a semiempiricalestimate of charges is warranted for a more accurate simulation ofthe molecular mechanics (dynamics) of polymer systems. The nextthree equations relate to the intermolecular forces found betweentwo (bond), three (bond angle) and four (torsion) atoms. The firsttwo interactions of this set are cast in the harmonic oscillatorformulation and do not allow molecules to dissociate. This has theadvantage that one can raise the "temperature" of the system ( asdefined in the last equation on the previous page) to severalthousand Kelvins and preserve the molecular integrity of the systemunder study. Not that one supposes that ordinary organic moleculeswould not dissociate, but this allows for the search of other equilibriaconfigurations of the molecular system under study through themathematical device referred to as "simulated annealing". By"heating" the molecular system to very high temperatures, one canpush the molecular conformations past very high potential barriersand sample other potential minima. This allows a more thoroughsearch of configuration space when looking for a global energyminimum. The torsional potential function is definable for all sets offour atoms which are chemically bonded. A simple cosine seriesallows the description of torsional barriers and wells. In ethane, forinstance, there are 3 trans and gauche conformations about thecarbon - carbon single bond and hence the value of "n" in the cosinefunction is set at 3. One could add multiples of the fundamentalvalue of "n" if different trans or gauche positions are not equivalent(as in the case of 1,2 dichloroethane). For a rather general andthorough treatment of the development of molecular dynamicsprograms, one is best referred to the Advances in Chemical Physics1

series. Although the title indicates a specialization in biopolymers,the background chapters are quite general and the extension toindustrial polymers is obvious.

I C.L. Brooks III, M. Karplus and B.M. Pettitt, "Proteins: A TheoreticalPerspective of Dynamics, Structure, and Thermodynamics", Advances inChemical Physics, Yoluime.. LXL, 1988.

58

* Annealing in Polyester Liquid Crystals

Up to this point there has only been a ground work descriptionof why and how to apply molecular dynamics calculations to polymersystems. It is time to address the problem at hand for thisconference. An ideal candidate for study of the annealingphenomena mentioned in the abstract is the oligomer derived fromthe reaction between phenethyl hydroquinone and terephthaloylchloride. An oligomer of 4 monomer units is shown below (figure 3).

Figure 3. The above oligomer is comprized of 4 ester units ofterphthaloyl chloride and phenethyl hydroquinone.

There are several key features to note on this figure. The phenethylgroups are relatively free to move about from a position of justabove a phenyl ring in the backbone of the oligomer to otherbackbones or phenethyl groups on adjacent, parallel chains. Thesephenethyl groups can readily assume an orthogonal orientation toother aryls found in the vicinity. This orthogonal orientationalpreference is demonstrated in the crystallization of benzene 2 and has

2 (a) D.E. Williams, "Calculated Energy and Conformationa of Clusters ofBenzene Molecules and Their Relationship to Crystalline Benzene", Acta Cryst,&A36, 715-23 (1980); (b) N.L. Allinger and JH Lii, "Benzene, Aromatic Rings, Vander Waals Molecules, and Crystals of Aromatic Molecules in Molecular

59

been incorporated in molecular mechanics parametrization schemes.The primary effect is that due to the acidity of the hydrogens on onebenzene ring being drawn to the aromatic pi electron cloud of aneighboring ring. This type of relatively weak interaction issignificant in the solid phase and when accompanied by many otherlike interactions, one gets a large total effect.

Another point of reference in figure 3 is to note that on theaverage there is one carbonyl belonging to an acid functional groupand the other carbonyls belong to ester functional groups. If onedoes his or her arithmetic correctly, he or she will see that there is aratio of 2n-1 to 1 ester to acid carbonyls where n = the number ofpolyester units. If one performs an FTIR (Fourier TransformInfraRed) spectrum of the polyester and fits the carbonyl peaks tothe two different carbonyl stretches, the areas under the curve aredirectly related to the ratio of 2n-1 to 1. One can thereby computethe value of n and the average molecular weight (MW) and averagedegree of polymerization (DP). The value of n is currently thought tobe about 10. If one assumes that the spread of the number of unitsper polymer is not very wide (which is typical for condensationpolymerizations such as this one), there is a simple equation thatrelates the fraction of conversion (p) to the average DP 3.

DP- 1-p

One can easily verify that an average DP of 10 would result in theconsumption of 90% of the reacting materials.

In order to compare the intensities of two different carbonylgroups, one should correct the intensity of the two differentcarbonyls so that they properly reflect the true concentrations of thetwo different peaks. The intensity of a particular absorption isproportional to the change in dipole moment during the vibrationalexcitation. This comparison of dipole intensity can be obtained usinga semiempirical scheme to compute the charges on each center(noted earlier as important in the electrostatic intramolecularinteraction). The scheme under consideration presently is thatproposed by Rappe and Goddard 4 . This not only allows for thecharge calculation at the atomic sites, but also an analytic expression

Mechanics (MM3)", . 8.., 1146-53; (c) E.G. Cox, F.R.S., D.W.J.

Cruickshank and J.A.S. Smith, "The Crystal Structure of Benzene at -30C", Pro&Royal Soc London, A247, 1-21 (1958).3 Malcomb P. Stevens, Polymer Chemistry, An Introduction, 2nd Edition,page 15, Oxford University Press, New York (1990)4 A.K. Rappe and W.A. Goddard III, "Charge Equilibration for MolecularDynamics Simulations", L b .. .Chem, 25, 3358-63 (1991).

60

of the change in charge with atomic displacement. Using thatinformation along with the normal mode analysis, one can computethe relative dipole moment chnges (intensities) for the two differentcarbonyl-. For an intuitive sketch of this method look at appendix Aof this paper.

If one looks further at figure 3, one will note that it would beinteresting to substitute the oxygen in the ester linkage with a N-Hgroup to form the amide to see the effects of substitution withheteroatoms of similar electronegativity. This results in thesynthesis of Kevlat, the tradename given to condensation reactionbetween terephthaloyl chloride and 1,4 diamino benzene. See figure4 (next page) for a pictorialization of three short segments of Kevlarsitting side by side with the backbone rings parallel to one anotherand notably there is the chance of hydrogen bonding between theamide hydrogen of one oligomer and the carbonyl of an adjoiningoligomer. This hydrogen bonding is expected based on the formationof hydrogen bonding found between beta pleated sheets of proteins 5 .

Experimental

To emphasize the interaction of undergraduates in this project,this section will deal with experimental approaches to this problemthat are under investigation. Undergraduates best benefit from anexperience to which they can ascribe some ownership. I have foundthat undergraduates can more readily appreciate interpreting resultsthat come from turning the correct knobs and mixing the rightreagents. The concrete operational skills leave them with more of asense of accomplishment and understanding. Providingundergraduates an opportunity to perform research is key toencouraging them into scientific careers.

Resuming with the thrust of .his proceeding, there are twoinstruments that play a key role in structure analysis that areavailable a: Butler. The first is an FTIR, which has been mentionedabove in connection with DP. This is pictorialized below

-0-OH vs --- OR

1600-1700 cm-1 1770-1800 cm-I

5 Lubert Stryer, Biochemistry, 3rd edition, page 28, W.H. Freeman andCompany, New York (1988).

61

Figure 4. Hypothetical stacking of Kevlar oligomers with hydrogenbonding interactions shown in analogue with P-pleated sheets found inproteins

Another experiment that will soon be possible is the use of acatalytic chamber which allows the infrared monitoring of solidmaterials at reduced or elevated pressures and elevatedtemperatures. One can perform diffuse reflectance on a sample heldat 5-10 degrees below the melting point for an extended period oftime to observe infrared changes that would hint at structuralchanges associated with annealing of annealable LCPs.

The second instrument available to us is an NMR which can beused to test chemical environmental changes for different protons or13C or 19 F. The 19 F would require the synthesis of the appropriatefluorocarbon to allow the specific site to be labeled for NMR work.An interesting alternate study would be Magic Angle Spinning NMRwhich would look direct at the solid state material. Otherwise it is amatter of dissolving sufficient oligomer to get sharp enough peaksfor analysis. Here again one could observe the integrated intensity of

62

the aryl groups relative to the acid proton or the hydroxyl protonfound (on the average) at either end. If each unit is the same as allothers (i.e., there is not a mixture of copolymers), then the number ofaromatic protons times the DP would be equal to the ratio of the arylintegrated intensity to the acidic or hydroxyl proton.

Theoretical

Currently, there has been the acquisition of a Silicon GraphicsPersonal IRIS workstation along with the software CHARMm andPolymer Dynamics (Polygen Corp. is the software vendor for theseproducts). Along with this the purchase of MOPAC (a semiempiricalmolecular orbital software package) has been made and thatprogram has been installed and tested for bugs. Last but not least isthe writing and testing of the code for the calculation of electrostaticpoint charges associated with atoms in a molecule according to theprescription of Rappe and Goddard. The formalism and code for thedependence of charge on atomic coordinate displacement has beencompleted. This when implemented into code can be used with thenormal mode coordinate analysis generated by MOPAC to producethe relative intensities of infrared fundamental transitions. The IRISis setup and the time consuming process of learning how to operatethe software as well as discovering all the software options is now inprogress.

Concluding RemarksOne might ask, "Could the (global) energy minimized structure

be obtained by starting with an arbitrary structure and usesimulated annealing to produce the structure of minimum energy?".To answer this question, one might first consider obtaining the a-helix structure of a protein from a random coil configuration.Consider a protein consisting of five amino acid residues (anunrealistically small protein) which starts in a random coilconfiguration. Proteins structures are characterized by two torsionalangles as shown below:

663

I IIH 0

Figure 5. The two torsional angles that need to be described per amino acid residue are indicated bythe curved arrows above. This unit represents a monomer of a biopolymer. There is essentially notorsional motion about the C--N bond around body temperature or lower.

The reason to pick 5 residues is that there are 3.5 residues perhelical turn, so the ability to recognize a helical structure requires atleast 5 or more residues. The final ingredient in this analysis is tonote that there are at least two energy minima per torsional angle.Therefore the two torsional angles per residue times the fiveresidues leads to the number of configurations of energy minima as210 (i.e., the number of energy minima raised to the total number oftorsional angles) which is approximately 103. That is a large numberof configurations for a small number of residues in a protein,especially when the energy search proceeds in a random fashion. Asone can see, it is fruitful to at least narrow the field of configurationsby some judicious choice (i.e., chemical intuition) of configurations tostudy. This does not even address the question of whether the globalminimum energy is the only structure to be investigated. Proteinstructure has been determined over many years of manyexperimental and theoretical approaches to come up with somecriteria for reasonableness of structure. In the final analysis, it is thesynthesis of theory and experiment that produces a coherent pictureof structure and reactivity.

64

* Appendix A

I. Electrostatics (Rappe and Goddard)

Single atom dependence on charge:A2 2

8E 1I 2 EEA(Q) = EAO+ QA AO + 2AS-'2Ao

or

EA(Q) = EA + XAQA + 2"AAA

where XAo is the electronegativity of element A and JAA° is the self-coulomb potential.

Applying this to molecules and allowing for chargeequilibration through coulombic interactions:

EQ(Q1....QN) = A(EAo +XAQA) + _ QAQBJAB

whereJ (R)-0J asR--0AA AA

where JA., is the two electron-two center coulomb integral which isanalogous to Coulomb's law except that the point charges arereplaced with electron probability densities which must beintegrated over all space in order to include the entire electrondensity. Ultimately, the above equations coupled with the condition

that the total charge on the molecular species equals the sum of theindividual atomic charges are reduced to a matrix equation whichdepends on evaluation of the two electron-two center coulombintegrals, and input of the atomic properties of electronegativity andthe self-coulomb energy. Solution of the matrix formulation yieldsthe atomic charges in the given molecule. This method has beenwritten into FORTRAN code and currently runs on a VAX at ButlerUniversity. There is under development the writing of the code forthe assessment of the first derivative of the charge dependence withrespect to change in the atomic coordinates. With this informationand the normal mode coordinate analysis of the vibrational modes ofa molecule, one can compute the relative intensities of all thefundamental vibrational modes.

S65

THERMOTROPIC LIQUID CRYSTAL POLYMER IMAGING

USING THE ATOMIC FORCE MICROSCOPE

Several thermotropic liquid crystal polymers were studied using

the Atomic Force Microscope (AFM), one of several scanning-probe

microscopes introduced by Gerd Binnig and Calvin F. Quate of

Stanford University in 1986. Each LCP is observed along two

sections of its material, one being immediately below the exterior

portion of the skin, and the other being within the interior or

core region. Samples from each section is cut to approximately

2mm by 2mm dimension and then taped onto a magnetic disk which

rests on the AFM scanner. The AFM, a Digital Instruments Nanoscope

II product, generally tracks over smooth surfaces only, and the

flow lines of each sample were oriented perpendicular to the

direction of the scanning tip. Images are obtained using a

silicon nitride tip, attached to the end of a triangular shaped

cantilever that is either 100um or 200um in length. A laser beam

at 670nm reflects from the back of the cantilever foil and focuses

into a photodiode sensor. Cantilever deflections, as the tip

scans over surface topography, cause the laser beam reflecting from

the cantilever to deflect. These deflections are measured by

changes in the light falling on different parts of the photodiode.

The AFM images are in the 'force mode', meaning that the 'z' scale

is in nanonewtons. The AFM process is time-consuming due to the

rough nature of these samples. Successful images, however, were

reproducible and the scan rate was no more than 3.5Hz.

66

Unlike the Scanning Tunneling Microscope (STM), the AFM senses

non-conductors, as well as conductors or semi-conducting materials.

The tip and sample are brought close enough together such that

electron clouds between the two repel. This electrostatic

repulsion is responsible for the cantilever deflections as the tip

"drags' over the surface. The atomic forces involved are of the

order of 10-9 Newtons.

Incremental movements in the 'x', 'y', and 'z' directions are

made possible by rigid piezoelectric tubing which acts as the

scanner element. Piezoelectrics have the property of exhibiting

mechanical strains, e.g. expansion and contraction, when subjected

to an electric field. The 'z' movement is controlled by a feedback

loop which uses the deflected beam as an input parameter. Voltages

along the 'z' direction vary in response to these deflections.

Illustrations of AFM images of HX-4000, SRT-300, A950, and

SRT-500 are depicted in Figures below. In each case scan sizes are

5um in dimensions. Skin and core regions appear to show structural

differences; the skins contains smaller nodules and lumps than

their corresponding core regions. The latter exhibits large

faceted domains more frequently than their counterparts. These

observations are in agreement with x-ray results.

The AFM images of HX-4000 for both skin and core regions show

the same orientation. A significant difference between the two is

the crystallite size. Average nodules on the skin are lxl elliptical

S 67

configurations that average 300nm in width along most of the

surface. The core, however, shows elliptical regions that average

1000nm in width.

AFM images of SRT-300 reveal high orientation for both skin

and core surfaces. Small crystallites on the skin region are lxl

ellipses that average 300nm while those existing on the core region

appear to be 1OX1 ellipses that average 400nm in width.

The polymer, A950, reveals a skin region that is highly

oriented along the flow axis. Nodules tend to be lXl nodules of

200nm width. The core region tends to be isotropic and containing a

multitude of large crystalline regions of 1OXl geometry averaging

1000nm width. These regions are often oriented at random.

The companion polymer of SRT-300, SRT-500, reveals the most

pronounced differences between the skin and core regions. Small

crystallites in the skin region appear as 200nm, lX1, ellipses.

The core region reveals larger crystallite structures, averaging

400nm in width and grown as 1OXi ellipses.

~~8

Schematic of Atomic Force Microscope

LASER COLUMATOR

LASER BEAM PATH

PHOTODIODE SENSOR MO

TIP

SCAN'NER

69

LIQUID CRYSTAL POLYMER DAVID S. SILVER

,a

ANNEALEDAFMTASK

CO G•MCIAIS HIGH MAGNIFICATION

UWANN'EULE

SYN•EISIs

1 - CO"CIAL - SYNTHESISY - MAGNIFICATIONZ -a ANNEALING

SIT500A950C130H.X4000SRT300

GRANLAX7

70

T LM e -t-inl 2

jgkldgtGjrýOUP

> 600

10= 340

0 0-

0 0. 0.4

0. 5

00

0 0~j O .~ 0

3it

iIeltifl& Point vs Structure.

71

0

o -

C.) C.)I... I-.0C.)

-o

00

C) 'V..

-� C.)

I-C.)

I-.C.)2

-

_______________________ 0I,-)0��

-0U

-� 0.-

Uo 0

C.)

2

V12�

Surface Spectroscopy

J. Adin Mann, Jr.Department of Chemical Engineering

Case Western Reserve UniversityCleveland, OH 44106

This project is just being funded so that work reported herein illustratestechniques rather than represents work done under contract. A major thrust is to usesurface analysis techniques to study annealing effects in the surfaces of liquid crystalpolymers.

The techniques discussed include the Raman spectroscopy of ultrathin filmswhich in this case are monomolecular films on various substrates. These surfaces arealso being studied by surface sensitive x-ray diffraction techniques. We are planning todevelop various ellipsometry techniques for the project with a special focus onellipsometric spectroscopy in the IR region of the spectrum.

There follows a set of figures that are briefly annotated to show results obtainedto date with sevei al of these techniques.

7

74

'."Lsrd assel/, iR-Y

II-

jt.

! A

I. Surface Modification

Surface modification by self-assembly of organic monolayers are shown for adetailed discussion of these methods see A. Ullman, "An Introduction to UltrathinOrganic Films .... Academic Press. ISBN 0-12-708230- I, QC I76.9.073U44,199 I.

?'5

H

H(S2) 12

&)OLi+*Before polymerization, the monomer inside [ • has two acetylene groups

separated by one carbon-carbon bond. The monolayers of both form solid, two-dimensional crystals. Polymerization is done in situ using UV light. The structure isstudied with the monolayer on the surface of the water, then it is transfered to a solidsubstrate using the Langmuir Blodgett technique (LB).

076

I=so40 -"

4n -0, -.2 0• --.80. .3 - -- -3

I 30 - - -

25 -- - -- - . - - . -

5-

F adow de~ws (nM*2/nM~bisl*

The isotherm of unpolymerized diacetylene. Polymerization is done with adilute LiOH solution as the substrate and the film compressed to about 15 mN/m.

777

Barriers

~Trouch

S~(Top View)

-,.ýI V;ndow

Rough diagram of the Langmuir through used to control the monolayer duringthe Raman scattering experiments. A Dilor x y Raman spectrometer was used; thetrough fit on the microscope stage that is part of the instrument.

78

II.|A

Si l -...- . ...

2,100 . SAW 1,100 60

Raman Spectrum of the unpolymerized monolayer spread at the air waterinterface. The bands have been assigned based on the monomer structure. Note thequality of the spectrum even though the scattering crossection of a monolayer is small.So far as we know, spectra of this kind have not been reported before this work.

779

Dl Wtin Monolayer-

.. ; ......

2.100 OI 118 0

The polymnerized diacetylene monolayer. Notice the simplification of the spectrum.

s0

DIACETYLENEMONOLAYERS

Polymer Monomer

--- .........

2,100 1,800 1.100 W00Wave Number (11cm)

Notice the sharpness of the polymer bands. The monolayer is a two-dimensioncrystal and is probably of large size (> I mm).

p81

LB film of a three step disposition of a condensed monolayer. (Shih, Johnson,

Mann - unpublished) AFM microscopy of LB films.

82

IGlancing I-lay Diffraction

GLANCING X-RAY DIFFRACTION: Incident monochromatic synchrotron x-rays are reflected downward onto the surface at approximately 0.50. Scattering isdetected at various vertical and angular positions by a position sensitive detxctor.8

83

SYSTEM SOLP TOP VEW

lbb

SYS1Th SE" W MEW*

Setup used on x23B, Brookhaven. The beamline is run by NRL-ONR.

84

30

110I

iU

The MTeta scan of polymriezed dicetylene. The spread monolayer way

polymerized in situ. The same technique was used as in the Raman studies.

I"

m • • 85

p -w

_ ' b.

Surface x-ray scattering can be done on smooth solids of a wide variety. Thismonolayer is an example of a self-assembled monolayer built up on a Si substrate. Thesample was provided by A. Ulman of Eastman Kodak.

86

".5 p . . .. .

2

+ b

*1e . .... .5 .- ' 1 -''-

aAVI

A scan of the reflectivity as a function of the incident angle (and thereby thereflection angle). Notice the well formed peaks and the sholders on the 2nd and 4thpeaks (= 5 deg and = 10 deg). The sholders are missing on the 3rd and 5th peaks Thiscan be modeled as shown next

887

• (OOL)

The scattering function for the layered structure was modeled and fit to the

data. Notice in particular that not only are the peaks fit but the details of the $holder atos

(00,.5A Icorspndn •7 e ae eroucd

2@

N8

St14

±*o.tsA MSo~A liA+0.107A W.3 1.2 •A

05 ooA 30.78 - -A

This figure shows the spacings computed from the model by a least-squarescomputation.

889

LCP MEETING

Butler University, Indianapolis, IN

February, 1992

Chlorine Edge X-Ray Absorption Spectroscopy

R. W. Hoffman, et al., Department of Physics, Case Western Reserve University, Cleveland, OH

44106-7079.

90

Chlorine Edge X-Ray Absorption Spectroscopy

ABSTRACT

CI edge X-Ray Absorption Fine Structure (XAFS) spectra were successfully obtained for

the first time in ammonium perchlorate (AP) pressed pellets during August and October, 1991

at Beamline X 19-A at the National Synchrotron Light Source at Brookhaven National Laboratory.

The research team consisted of Professors R. W. Hoffman and J. A. Mann and Mr. G. A. DeRose

from CWRU and Drs. K. P. Chaffee and J. J. Rusek from Phillips Laboratory, Edwards AFB.

As is common, the Extended XAFS (EXAFS) spectra (greater than 50 eV above the Cl

absorption edge at 2823 eV) were weak and damped rapidly, suggesting that the coordination was

low-Z. A detailed phase and amplitude analysis of the EXAFS spectra is in progress and will

be discussed later.

The near edge (XANES) spectra (within 50 eV of the absorption edge) were stronger and

0 more intense. XANES expresses the Cl-local symmetry and multiple scattering and band

structure effects. Unfortunately, analysis is difficult as no reliable computer codes are available.

Our analysis is presently in the fingerprint stage.

Energy shift and spectral effects are presented for AP pellets with monolayer drops of

various binders; we have completed and are testing program changes to use the inadvertent air

Ar edge for energy calibration - a technique not readily available for the fluorescent x-ray

detection used because of the very soft CI edge energy.

We have also observed orientational differences from pressed pellets, and single crystal

AP by taking advantage of the highly polarized x-rays from synchrotron radiation sources.

Copies of viewgraphs used in the presentation follow:

091

SYNCfHROTRO%.N RADIATION

ELECTRON ORBIT E=ytncZ

ACCELERATION""- ARC VIEWIEDBY OBSEFRVE;-R

ev E

CASE 11:I

F~or 6-GeV Synchrotron:A

T(~11742

8% ~mcTLE. 1/ 85 rnicr'oredians

92

5. NSLS BENDING MAGNET SPECTRA

1 1 1 1 1 I 1 1 1 1 I1 1 1 1 I I-

0 0

V93

LUJ

cr 0 0cc -1ýA

9c4

00r

> -.

uS Ln 001

zcl-4Z

16*

la I

495

C CZ C 0'

amEvON~I-U A

WAA

Cc

I- j

C C

Ux 0in c

19 .~ E

U.4 -0 go

£ tow o C EO~W I. o w- C . 0 .

E 0cc S.. c

L c r

c w40

00 0

U~0.

96

SI

Characterization of the NSLS X-19A Beamline

A. Tunable ,nergy Range (keV)SS(11r) (1) 2. P - 7.9.1

(2) 7.40 iS.O

Si(220) (1,) 3.46 12.0.

(2) 12.08 - 22J

B. Beam Spot Size

Unfocused: 4 mm (V) x 40 mm (H)

Focused: 1 mm (Diameter)

C. Photon Flux (Photons/See)

2.5 keV. , 1010

5.0 keV - , 101'

10.0 keV- 5 x 10:1

at I=100 mA, E,=2.5 GeV, Si(111) c:ystal, unfocused beam.

D. Energy Resolution (eV)

1.1 eV -5i(111) with 4 mm slit at 2.5 keV']

L0'0.5 eV 'Si(111) w,--- 2 mm slit at 2.5 keN']

1.4 eV 'Si(111) with 0.1 Mnm slit at 7.0 keV]

0.8 eV [Si(220) wth 0.1 =m slit at 7.0 keV]

E. ?olar'ration

"-' at 10.0 1.eV amd __:-am slit

X-1 9A Characterizatlon

97

1.• • •.L•

uj

L En

10- 1 06 4

0

41 1

41__ _ _4

r 4a

41 4441 U U

- eft

L l14)4

98C ~ I

C6 K,*

Al,'

slof"L6 Sc~rlwkimt4o Ctzfc.UpTs

99

UU

= 3c

100

00

>b

.-. **%

_ p0ti

0 CO

a Qc-

U0

p0

Ecis=IL

1wocW

2C

101

25F

20 Fc-Fc

x15

lv ~~pieSr S*fU.. WA.INDOW

c 10-

0, -J5- ~I-

00 1 2 4 5 6 7 8 9 10

I I Radius (A)

Fourier Transform For Iron Metal

102

• •• .... .,•.••.,. ,.,,.;;• •.* • *-. ,•. ,; •, v

0

EXAFS XANES

(2)(3) /(3)

SAbsorbing atom -.-- Photoelectron pathways

'A:A.

S03.Wý Y.

103

AA

'!I " I I '1 I.11

364

T

0 21.1 21 o.258)

X10

311-*

o. 1

M ZMI 280. 21 294. 286. ZWI M9. M2 294. 296 ....

DEY (k'V) XIO

CL04NN4. 30PW/ 3/1 G-AUG-9 I F1g9'e 44 CLedgs scan for ommwgy calibration

RI.0 . .121.288

1.

0.

I

F 0.FE: G.

N 4.T

X30-1 -2. -ii

-,4.

2M 279. 20 . 23. 2W. 2 2M 4. 04. 202 24. 296.

RY (mV) x3O ICLO4NH4. 3 3-• " -AvJC sr,

6.98.XAPXTL. 192

0.20

6,.7

01.60

0.40

0.30

0.000.208.4

E - eo(eV)

Fig~ue 2: XAFS= Slpecmzzn of NH[ICIO. single crystal with pie-edge zuuovud ni o E, ,,2722.7 eV.

. .5. .. . .. . -.. , A, -•

AP. 1088

2. 20.

1 .S56

1 .66,

6.153

-82 50002 .46 6.66 6.96Xle 3

E - E. (ev)

Figure 1: XAFS specmum of NHCIO, presued powder pellet with pre-edge removed relamieto E, = 2722.7 eV.

10

106

ii uti m h. 1,8 A m

N

jCY

ea

N ci

N I

107

a I.+

Noco~

.-

6ww

00

06~

SE

-I- 00

0 w

.1106

+

0

+z

oc

(.wn oft* 0ANAo-j 9g

+ ja

+

0

z

so CL

4a

+w

+01w a

0 o >

o 0

(dunfl So~p) I(A*,I.A-ot-Ivo3d ItL)

110

Autust, 1991 XANES Results: Sample Side Dependence

Sample Normalized Jump.................... Io • ; • B ................................ ............o.o........ •................................Top Towards Beam Bottom Towards Beam

AP/Toluene 0.645 0.644

AP/TET 0.5 ML 0.650 0.641

AP/TET 1.0 ML 0.642 0.624

AP/TET 2.0 ML 0.588 0.480

AP/Blank 0.650

October, 1991 XANES Results: C! K-edee samples

Sample Normalized Jump

AP 0.650

AP/CLHQ 0.440

AP/PECP 0.213

VCl3R 0.131

VCI3 0.185

KCI 0.248

NH4Ct 0.272

NaCl 0.408

RbC1 0.532

SYN1I 0.572

SYN1. 0.516

SYN13 0.585

SYN14 0.544

S111

(dwflr oBp3)ldojp KOIINA-)IuSd l8I.W

CeO 46 csc 6 6 06 ci 6 0i1

0) +

N0

00

60 zz IL00

+ 0.+ I IL

(A*) Wd) *3 - f1U! dV) *3

112

1992 APC Symposium

INTERNAL REFLECTION FLUORESCENCE OF POLYMERIC SURFACES.P. B. Oldham, Department of Chemistry, Mississippi StateUniversity, Mississippi State, Mississippi 39762.

The characteristics and dynamics of materials in the liquid-crystalline state have intrigued both scientists and engineers fora number of years. A considerable amount of work has been done tocharacterize the bulk properties of liquid crystalline materialsbut there are significant questions concerning interfacialmicroenvironments which remain largely unanswered. The goal ofthis work is to combine the capabilities of site-selectivefluorescence probes with the surface selectivity of total internalreflection fluorescence (TIRF). Through the use of opticalwaveguides, TIRF can selectively probe the first few hundrednanometers of a surface interface. This technique will thus beemployed to investigate liquid crystal polymer (LCP) materials ata surface interface over a range of temperatures. Probe moleculessensitive to both the physical and chemical environment will beused. The data obtained should provide considerable insight intosurface interactions involving liquid crystals.

The major thrust in this research effort is to determine thefeasibility of TIRF spectroscopy to monitor LCP deposition onto asolid surface from a solvated environment. This entails an overalleffort including the following tasks: 1) investigation of solventsystems and solubility determinations for commercial LCPs, 2)chemical attachment of fluorescence probe molecules either directlyto the reflection element surface or to the polymer itself, 3)optimization of optics and investigation of surface growth kineticsof the select LCPs using TIRF.

1113

0

INTERNAL REFLECTION FLUORESCENCEof POLYMERIC SURFACES

P.I.: Dr. PHILIP B. OLDHAMDEPARTMENT of CHEMISTRYMISSISSIPPI STATE UNIVERSITY

0GRAD. Mrs. DEBBIE BEARD SAEBO

0114

PROJECT OBJECTIVES

(1) POLYMER SOLUBILITY

(2) CHEMICAL DERIVATIZATION ofPOLYMER SURFACE

(3) MONITOR KINETICS ofPOLYMER DEPOSITION

115

01

'

CO

00Cki-

o 116

S o= Sin -1 (n2/n )

Normal

b etotally reflectedC - beam

10,*

refracted beamn, n.a • Incidence angle c critical angleb * Incidence angle ý critical angle

S117

Normal

Incident beam totally reflected

elg beam

\nevanescent wave

"--z 1 dp

118

d (0 1) U • v "• ). - -

Oi a incidence angleXi a incident light wavelength

n, and n2 m refractive indices (n1>n2 )

a

ION, k, fl, f C~e (-,S"".,) dz0

im =km/e(-sldl d'

o -

I k.,k,.W5zif C~e" dz0

119

ToW hibimd R~m Fumag

by

30

~25

x

4.

v 0

0 0 40 s0 80oU

Incidence Angle

120

OD

0.

=WN

100

121

Ftc.~~~~~~~~~~~nl lA h eg~*vral-nl elcinacSory.

1222

Possible TIRF Experiments

I. Fixed Angle Surface Polarity

2. Variable Angle Depth Profile

3. TIRF - Anisotropy for Mioroviscosity

S123

SO

CL

bo•

0S

0 00

C1

o oI

"0'

Cb

0O

I 21O

SPyrene

I~lll

S M

(AA

Figure IV-11. TIRF spectra of pyrene in cyclohexane and inmethanol. (Emission wavelength : 360-460 nm).(X-axis : 1 unit - 20 nm).

1

'12

10

0)

tol

10

(0

to co) 0)N ~ W0 (ID

126

Is-

00 ~CM

w200

wzw(0

Go

U. CM

Om 0

oc

V- 0 ;c;c

C 12VL

Analysis of VA-TIRF Data

Define Function

Input: Initial Estimated AdjustParameters

Simplex Iteration"To Find SatisfactoryInitial Estimates for

Least Squares Minimization

Least SquaresOptimization

Output: Best FitParameters

128

Fluorescence Anisotropy

(r(

Rotational Correlation ViscosityTime

1 +kr/4)) RT

PO = limiting anisotropy R = gas constant

T = fluorescence lifetime T = temperatureV = molecular volume

1129

x

I!

00

-30 .

130

03

a CL

U3 0

03 =i

* 131

100% PARAFFIN OIL

a- W- D

VISCOSITY (cp)

132

0.120

0.100 A

>.0.080 A0. Ca0 IlL Aif//

* Q ~0.04o- I•' - Unmodified0.020 0,

0.000 i - / "0.0 5.0 10.0 5.0 20.0 25.0 30.0 35.0 40.0

Viscosity (cp)

133

SOLVENT REQUIREMENTS

(1) REASONABLE POLYMER SOLUBILITY (few mg/ml)

(2) RELATIVELY SAFE AND EASILY HANDLED

(3) OPTICALLY COMPATIBLE WITH TIRF

134

SOLUBILITY PROCEDURE

GRIND POLYMER

DRY POLYMER -36 HOURS

WEIGH -5mg POLYMER & PLACE IN VIAL

ADD 10ml OF TEST SOLVENT TO VIAL

COVER VIAL OPENING WITH ALUMINUM FOIL

LEAVE IN DARK AREA UNDER FUME HOOD

SHAKE VIAL FOR -2 MINUTES 10 TIMES/DAY FOR 18 DAYS

ALLOW UNDISSOLVED POLYMER TO SETTLE TO VIALBOTTOM OVER 24 HOUR PERIOD

PIPETTE SOLVENT AWAY FROM SOLID LEAVING -2m1

EVAPORATE REMAINING 2ml OF SOLVENT LEAVINGUNDISSOLVED POLYMER

DRY POLYMER FOR 2 DAYS AT -130°C

WEIGH POLYMER

SOLUBILITY CALCULATION:

(INITIAL WEIGHT - FINAL WEIGHT) mgTOTAL VOLUME OF SOLVENT IN ml

135

00

C6L

0L

'"1CUC.

4'

I-.

z

1.4 I.I

8 8 0,IWE WE 0 0

+ + + + +

3ONYUUOS9Y136

Room Temperature Solubility of HX4000 in different solvents.

SOLVENT SOLUBIUTY(mg/ml) ABSORBANCE XMar)

Acetone 0.070 ....

Acetonitrile 0.100 0.12328 244

Benzene 0.100 0.01759 282

sec-Butanol 0.100 0.00703 328032249 280034500 271036200 260

tertiary-Butanol 0.060 0.00580 2620.00642 256

Cyclohexane 0.110 0.47130 3121-34530 2761.53490 2681.50040 2603.22840 224

Ethyl Acetate 0.080 0.05356 360S0.35486 252

H exane 0 .080 ....

Methanol 0.100 0.01796 244

Methylene Chloride 0.180 1.44060 246

Tetrahydrofuran 0.130 2.61060 254

Toluene 0.050 0.05203 284

o-Xylene 0.150 0.02919 288

m-Xylene 0.100 0.03366 3260.22304 288

Ultra pure Water 0.120 0.01949 244

137

000

) i)a.•

a.0 m >. 1 wI.

0>0>

0 +LL0

aQ0 00

o. x

0.

138

SVS

iN

* I-

60 e

"*- '

C llSLZ ALVI

Oz

o1 9

Abstract

The simple and reliable single filament test methods

for predicting the compressive properties of fibers are a

must for development activities of fibers in laboratories

because of difficulties in composite compression test

methods. Therefore, in this study, two single filament

compression test methods, the elastic loop and bending beam

tests. are conducted for several polymeric fibers including

Kevlars, FBO and PBZT and a few carbon fibers such as T-50

and P-75S, in order to obtain their compressive properties.

Also the compressive failure modes of fibers, which occur as

a kink band formation in polymeric fibers and as a fracture

in carbon fibers are investigated. In addition, a FORTRAN

program is written for numerical analysis of non-linear

geometry elastica problems such as bending of a single fiber

considering large displacements.

A comparison of the results obtained in this study is

made with previous studies. It is found that,

generally.the compressive strengths of the fibers obtained

from elastica loop and bending beam tests,are higher than

the composite compression test results. The kink band

formation in polymeric fibers were investigated and it can

be concluded that the critical kink band formation

represents the buckling of separated microfibrils due to

140

P elastic instability. Alxo the FORTRAN program is applied to

measure the fiber compressive properties as a new potential

single filament test method.

14

I.

pi

141

Os

020 l

0- 0U owco cNL 0- c

-0 (u C)

L. C

como Cn-cc -

(D)

4142

C/)wC~O _ mDc

U.L LL

Z W L.

H>-

* Z

H~C/)w

2 w~0

X LL ZwO0(j

________________

C. 7

II

,0 -

o C- 0•

C. 0_C00 2oc 0 c .

- C " 0 r_

E 14)

OCDCD) 0) 0 &C L..2M coa L.

0- 70 >#700

0) 0 0)

at. oo _ EL

0 0(O

o.2o

144

sm E O

_ o0. .o

0.0

0- vl _ 0•cc7

CI . o) C°

>J CDW 0 -a 4mE

o 0E CL 0EI 0

EE

1.oc mdCJ

00.

, '0 ' C, ,E

0U

E co

0

E0 0

0 ~Co CCIor 0 . C c

4) 00 E2

W 0'4 0

Ca Co C). 3

0.. 00'4

D 0 L0

Omo U) U)

~.0 >E~UJ Co Co

ca Cocc) .-0 0M

0 0 0

146

OrientedFber

O -5 y--m• facro FRbrff

- --0.5 m Fibril

0.05 pm Micro Fibril

1 4C 7S--r

147

CO)L.

c0

0,._ 0

DL.) 0.

- -ge

• L. .-EM,,

0> 0 0Mco Ee I•co m E oa

L Cc Inn , .. .a

1(D

z

149i

iN7*

* * 9. ~ Zi

*. .- I

.9-",a

Co Co C =

CCc

0. or

*0 0 )

U0 00 (1)0 0.)

1- 0

_ 0

0 oH co a) Cl

W .a 0 E 1

0 E)

0 mm2o.

00 0-,*0 Cl 0.

H .0

(p4) L. 152

Co CL

l9 J

0 C9

0.... -04 0

0~44

10

CD153

1p N ,1

NN

p

9Cn E_2 °.

ow 0c

/E

J) Ec 0

cc Cl'> E .,.-

4,..

L M C=oz

04) a: ) 4)

0•_ 'm::C/ co o

wE-,- 00 .-1

Co "" (PSc: .4)4: ) .-

mm.CbLL

154

F- cnc

<t4 O

mm..J I-q

Vs Ef) - 4

".. , w. a C

0 0

E4 .. 4

I inQ g bC12 0 + U

* IL +1Ix

uas

0 155

7 1'CC)

0 Iwo_ (o0.E 0 -4 V -

0 42o4d S. * o .

0 C

4

j _ 0

4) riift

I 4

000

dI1156

Y3rI

i7

Cl)

F o- o

0.. V 0cz0 0)

04-0 4.0ICO) 0 C

o o ._cD

Cc L

EH

E E co 0

157

O0"0

0 co

0l E

H 0"0•.. 0 0

0)o 0"

CP0 0)0

0 CD

w01-

0158I _ ,, t

IH

H ii.

0' Iv,

mm-O! '

159

H V

0ý() L. S.

in) - un _

o4- co

=O 0)

W4) CO asd

co £1)m

-l j 0 a F-

O E60

70

CD) ma

L.0V

W I I 11" -Iw ) __o AL

(1) w -IW >E __m

oo (D

z > )* co

W0 HCO)

_ca " 0

Cc a c0

I l

w E Co >>)

z~ oEa) >0•o .C Z.X oi..

00CU 4~~) .ý

OL .CC) CP I-a 00)-.

0 ccco Zx zoa: *" '*E:

m00 T 0 %w0

c 16

162

w

0) m 0)

00 00)0m c .

4a)

Z Em-o-G 0 ;v,4

E "04ac0 Eca a E

co 4"

>0 E*E 0 '-Ea 0 o co~ m

0 0 E

163

w V

ccc

Ho 0sU- 00U

Z0 E 0 L o- C0

0S.J.- '- E ' -C C.-

0 o_ U. o

- mom4 ) E 0

CD va 0E, _ i ._0 0 0G E

c~~G) 0e0

00

164

w

ca)

~ca)

0) L.

wI

116

CO)

/V

"0

m~

Ooo.0 L.- o

00

x 1 V 4

Hc~%* 0 0m~QL00000ai M.0)c

>0 C.O . 0a 4" C

L.. A E~.

on- 0D W, 0C

16%k

Clo

z c) ) a)w EE o

-0 ca

Co O. CO

CTO Coa

167

A/•A17;,,g 3,

A~ 1A

4 ft'

The load is r'eleased'11A/

168

00

0 co

>a

o c1O 0 0 0 0 0

0

ca -0

0:E

~Cl, E

CD 0

LL

C,. r--rr--. C!0

169

pW 0

(/4) ___ * 9 9 *<) >

o LL I0)I0 9 N 0. 'a

U .Ii .

- -u- .- -- *-V

" 0 40 0 ># 4 .1

2- FOS WU C 0

co o1..1-'- Ia 0 Ol - a 0 00 1~ 000I r" 0 0 1 L.V) C -I

20 asd6

aU .. 0 x , hib 4 ~.. 4 >

n~ *0

a I- a

a o '0ui Lo0~ a

0 0 0 0 0 0#6. 0 0.0 S

C.)~~~ 0 0a90

oc r 6 0 0 t fL.1'.-4-0 4

Na 0 09 1

170

coz coO

0o E

Ix 4)

0 cono

000z 0

Cog )CD,i--. -) -vo s -n

4=0 -W CO

I1) 0a"'

cc

171

CO

00)cc DC

CIOV 0I U8 0

CC

0D o 0 icc0m G) CC~0

a.~1 -o co (DU

172

00)0

4) (MCl0 CIO_

(1) ) c0L. 4- L.0)L

<I C)> )OOco -W Lmas...

CCC0 ) 4)

(D c:.0)

co ... L ( 6ý 0

04) 4) 0

0. 93 >

Z co L.Co 0173

SCO0a0

0o &.D

v o

Q. 04- CtE

O 0 aL)0E

o O0 QU4COCU)) 4)~l

0 0 1ts.02O0

CuMCci

MEOW74

00

00

a C '0 0.2 E''

o 0)0 01)-0 co E L

ono

z0

* E

' S W

17.2

a) .. a0 0)

co CIOL4.a

Zo 0o > >a0 0..

o q-0 4)a)(1 4)E4 .C

000E.> co CO)> 4n)

(.) 0)' L..

00.

E . 0 >

176

CC

0 0>

0Ct)

E E000o1- (1 0.

Cl) ) = 4-00 c C c

0 ca

Co ~Co C0UO (. c)C

xo 000).0)C>~p Ci0CI0

(UEn-0

000

0m W0C 0 70*§ a

aIf

177

0 E

0(104 ~ca

MINO)c E

0 *.a

z CO C'4-C 04= 0

00 4) 0 2

-0

00

*a 0 0

0 ow118

co

co4-Lw.

C I.- .

C'))E "Imme

0. (1o E.- "CO O

ci) 0QCl) m -z 0 ) - 4

0 "- • 0 0 COLoa 0.

4 ) 40 "-- a -o 4- 0 W) 0..o *

>. 0

0 E ,oc) 0 -o._o Eo C

FD

179

z .~Q i

00I- COOU 0

onoZl ao:W=o -0

slw0

4-0 Zo

E

Cl))z cac

COO

IL0-

0 Ct0

C.) L. . .

0 0

181

Abstract

Characterizing the time-temperature dependence of themacroscopic states, the phase transitions and themolecular/morphological structure is essential to the developmentof high temperature thermotropic resins for low cost fabrication ofhigh performance primary structures. Frequency dependentelectromagnetic/dielectric sensing techniques (FDEMS) provide asensitive automated means for measuring the time-temperaturedependence of resin properties and changes in state continuouslythroughout the entire cure cycle. FDEMS has the potential addedadvantage of measuring these properties in-situ in the mold duringfabrication, thereby monitoring the particular time-temperatureheat transfer affects of individual molds and fabricationprocesses.

In this feasibility study, using DuPont's thermotropic liquidcrystal polymer HX-4000, FDEMS were shown to be a sensitive in-situmeans for monitoring the effect of time and temperature on thephysical state and properties of the HX-4000 resin. The sensorresults were correlated with differential scanning calorimetry DSCmeasurements which also was demonstrated to be a useful laboratorytechnique for characterizing these changes. The successful use ofsensors in-situ in the mold in a high temperature pressdemonstrated the potential capability of using the technique, bothin a laboratory and a production environment.

The sensor measurements monitored the changing fluidity of theHX-4000 resin and the mobility of the polar liquid crystal moietiesin the amorphous resin state. The FDEMS measurements indicatedthat in HX-4000 the resin anneals at temperatures around 300"C overa period of hours and that the resin undergoes further reactivityat temperatures near 380"C with time . The DSC results support andenhance these correlations as does the work on the effects ofannealing on the mechanical softening temperatures which isreported on at this symposium. Overall, the results of this FDEMSfeasibility study on HX-4000 strongly support the importance ofdetecting the changing properties of thermotropic resins and theability of FDEMS sensors, corroborated with DSC and mechanical workas a means of monitoring these changes in-situ in the processingtool, both in the laboratory and production environment.

182

Frequency DependentDielectric Sensor Measurements:

An lnsitu Technique for Characterization,Cure Monitoring and Process Control

David KranbuehlSean Hart

Yunfei WangChristina Short

Department of Chemistry and Applied ScienceThe College of William and Mary

Williamsburg, Virginia 23187 - 8795(804) 221-2542

1183

0

Immediate Objective

This year's immediate objectives are:

Characterize through DSC, RDA, and FDEMSsensing, the time-temperature dependence ofthe macroscopic states, phase transitions, andmolecular/morphological structure of high tem-perature thermotropic resins.

* Demonstrate the ability of FDEMS sensing toprovide a means for insitu monitoring of thesechanges during processsing.

0184

I

Overall Objective

To use embedded wafer thin frequency depen-dent electromagnetic sensors (FDEMS) for continu-ous, online, insitu measurement of thermotropicproperties, in the laboratory, in the fabrication toolduring processing and during use in the operatingenvironment.

* Cure process design and optimization• Closed loop intelligent process control* Life monitoring, smart materials

1185

Two Molecular Probes

Ionic - free charge ..•E _translational diffusion of ions

px -forcePTdt-

+ 0 -sphere: =6 6-"rT

Dipolar bound charge T

E E

rotational diffusion of dipoles*p dt torquePRdt

sphere: P 8 v7- r3

186

II

10 II"

+ +

187

• *. ' i~

0 04

b-d

o No La 3uzU) Li)

II 00

Li187

Q0

0 (EU 0oCLC) CJ

TC)

04

0 LCO

01

L)O

0c) (- L o '

-~ 0

000

+ 0

(4-J

000•0

LoO

0+ +r

C4dI -

000 0

---- *0o ,

S4.00

11 +9

• +

÷ 0 0o 0 0 0-

+ -+ +00

a) a) a) 0 00 0 0 0

+++0 0~

169

+

a'. -+

00

0 0o

0*IC)

÷0 0000

1-

@00

CO-LL -. L.)/.190

C00oI m0C)

UC)10U

Coo

0

C(LOc,

piCo9

~cnCDa)

= ~ ~ 0 VE

o- EE• l. 5.2.9---0

-.- o i -,"

CD

0M

0--c

D - E Wcc N

- - 0

E aE

w~ E_--) .. , Z-0.o= CCL 0 Co oQ- i

= Q~-j E o

CD. -0 0 r_•

(n.o I.*,,O@ @ -.Wa I 0I•C . •¢ -0o. . . . ...

0.c

U) 0 Lr

0

CU)192 •

Advantages

"* single inert sensor

"* temperatures to >800 OF

"* continuous uninterrupted simultaneousmeasurement of both e' and e,"

* 4 place sensitivity in ,'

* 3 place sensitivity in e,"

* permittivity range 10- to 107

° frequency range 10- to 107 Hz

° low cost multiplexing - multiple sensormeasurements

193

Initial Materials

HX-4000

VECTRA A950

VECTRA C950

XYDAR SRT-300

19194.

jainqejadwej

o 0 u 0 o 0 0 0 0 0u 0o (0 cu 0u cu C o 0 0

(T) C) CU CU CU '4 '

crI,

a)

CCD

(D _

Go (D c/M a01

195,

C')Vm

c-u

0) C-

cu 4-E"C- CO

f cu E

r.o-- C D

.0 cz0

C-C

SLD (DC) ' U U C

M*,, 901196

-7 C

CDr

V3 Cr)

a-a)c-'J

cu4~CD cDu

cuE4-J-

ca

'fIn

197)

L9

KLoI

41 I~

CD7

CD -dCD M

Im L

(D-e IQ Su

coc(K.

m~ dd ci0

cS -~C3

a Lnoo

In U' n t n toan 0ULd o~ IL~0

li- 198

3 ainqedadwajo cu o 0 0 0 0 0u0oT D CU u 0u cu U 0 T 0

Sm

0) 0I- (I

CU

U)

r-1-

4-Qco0

CL Coa, .14

If)If 10 wi n .4 IV m m cucu

(M* 9011994

3 ain;2Jedwajo0 0 uODq 0 0 0u 0 0ow co Cu cu cj V cu o 0q C

to

CYI)

C) in

4-J

/I. cu

fn 0L:OO

C3 .

U I OD

U)o

vam ý c.20

3aun;2Jedwa81o o 0u co I 0 0 0m 0 0oV co 0 cu 0u 0uw

0

Cu to

c-uCD

-D c

CU

(10

CLU

(M* 00

201

3einqejedwalo 0 0 0 0 0 0 0o 0 Nm 0 IV 0 Nc 0 0IT) m m cu cu cuw V

0m

CC"(

C~CDcu

0,nCuu

cu 0

LO)

cnE

c~c)ro 0

200

IIQ0j

a)

V-46-4

LcuU

a)

ciii

.06

0 to

0

0..

(sun it

*203

[C0 L)Kj

cr¼

w QoCL-Cc;

.0 u

w m

Wa; ; n

OJ 000

Cc M

X cq

-0 07'-4 PiunCj 4d c

0m 9 C ZC NL

a 0

N0 L** .

C) (A)UllW0

I.E

2d

- a-

- I-

r4 4w cm01C

cy C

M' c-nsc c

f U)U

ICU

4-)

LLJ-In

Il LCD C

W La- -Cc

cooLLu

ClCa

da d0 m

00 -: A 0 W

I* NY 0 N wj -V d 6 w to 6

.0 'n00a a a v

15 - x

t- 206

ADVANCED POLYMER COMPONENTS

RHEOLOGICAL CHARACTERIZATION

TASK 8

DANIEL SCHWARTZ

PHILLIPS LABORATORYPROPULSION DIRECTORATE

EDWARDS AIR FORCE BASE CA 93523-5000

ABSTRACT

Rheological characterization of polymers is crucial for optimizing material selection andprocessing conditions, but it can also be used in identifying phase transitions and predictingmolecular behavior. Dynamic mechanical measurements conducted over a range of frequenciesand temperatures give the materials viscoelastic behavior and dependence on such variables asstrain, temperature and frequency. Therefore, insights into the polymer's microstructure andmacrostructure may be obtained.

The material property of greatest interest to the Advanced Polymer Components programwas the annealing phenomenon exhibited by some LCP resins. These so called annealed resinswould show higher strength, chemical resistivity and thermal resistivity in response to a thermalconditioning above the glass transition temperature but below the melt temperature. This wouldbe detectable by observing an increase in the storage modulus or a decrease in the dampingfactor.

OBJECTIVE

To determine the viscoelastic properties of the LCP resins selected for tle APC program.This includes the identification of phase transitions and predictions of molecular behavior. Ofparticular interest is detection of the annealing phenomenon in those resins where chemicalcomposition allows such behavior.

INTRODUCTION

If dynamic measurements are made at a fixed frequency over a wide temperature range, theAlpha, Beta and Gamma transitions may be identified (figure 1). Transition zones may alsodetermined by conducting a frequency sweep over several decades at a temperature betweenthe glass transition temperature (Tg) and the melt temperature (Tm). The Terminal, Plateau andTransition zones give indications of molecular behavior (figure 2).

Changes in the dynamic moduli are sensitive to materials becoming more stiff or soft, thisallows for detection of annealing.

207

~NVI 1-

o0 0 0 C00 :0

I~lm0 <u

W 0cow

0 -J:

Lu 0 0 Cl)j:~ a 0

0.0,1 ~%o

0_ :)- i

0 <<0 0~

02 0l

IUj.6 - -0

Z 0

TOO 0

0 0o

208

[ s Sfl fl;XQ SSo 50

C3i ~ I9339) )2 Iu I il

LUj

LiD

L t

0u 4 0

I-

**0 **LL0

L E

20n! I U.- 2

1 7%

EXPERIMENTAL

The viscoelastic properties of the LCP's were measured on a Rheometrics MechanicalSpectrometer model 605.Viscoelastic properties are determined by subjecting the sample to a sinusoidal shear history(Dynamic Testing) and measuring its stress response. An entirely elastic material will have asinusoidal stress response in phase with the inputted strain and a viscous Newtonian material willhave a response 90 degrees out of phase. A phase angle lying between these extremesindicates the material is exhibiting a combination of these behaviors. By measuring the stressresponse of the material and phase angle, complex moduli can be determined (G*). From thisthe dynamic viscosity (rC*), elastic component (storage modulus, G'), viscous component (lossModulus, G") and damping factor (Tan Delta, the ratio of energy lost to energy stored) may becalculated (figure 3). G', G", and Tan Delta represent the viscoelastic properties of the materialand are functions of Strain, Temperature and Frequency of oscillations (Rate).

For our tests conducted on the liquid crystal polymers, frequency sweeps were run from 0.01to 100 rad/s (.0016 to 16 Hz) on a Rheometrics Mechanical Spectrometer model 605.The samples were measured in parallel plate geometry at temperatures above Tm totemperatures approaching Tg. The properties measured were G', G", and Tan Delta. Theseproperties relate to a polymers ability to store or dissipate the energy of the deformations(strain) applied to them and are influenced by the molecular structure of the polymer and the testconditions.

The first frequency sweep was carried out on a sample of HX-4000. At the lower frequenciesthe viscous component dominates with G" values being higher than G'. The material is in theterminal zone which gives indications about the relaxation times of the polymer molecules, Thiscorrespondes to the materials molecular weight and molecular weight distribution. The pointwhere G' crosses over G" begins the plateau zone where nearby molecules begin interacting withone another. The amount of entanglements and crosslinks are obtained in this region. Becauseof the inverse relationship of frequency to temperature, as the test temperature is raised, thecross-over point is shifted to the right. The entanglement region occurs at higher frequencies dueto increased mobility of the molecules (figure 4). For HX-4000 we are unable to go to highenough frequencies to see the next cross over region which would be the transition zone wheremotion is due to branch point and group motion coming off the chains.

210

LU

'0 0

C-0

II U) II It a

* z1

V) U) _ _ _ _(

nloOiý ol le'FINV I

UJ

211

P) m

cc '0

00En a '4

CI'

U, It1C3q~

w 1CCI

0 ¶

L.w

I-

-J -

ccE.

(n

0

xI0

0 ) [3Uo/UApJ .9 (V) [pUo/UAP) .9

212

For Vectra A950 we observe the cross-over points occuring at higher frequencies and areable to detect the start of the transition region for the sample tested at 31 C. Vectra appears to

have more molecular mobility than HX-4000 and would account for the lower overall modulusvalues (figure 5).

The next test was to try and detect the annealing phenomenon in a sample of Granlar resinwhich had been heat treated for three hours at 250"C. If annealing is occuring there should bemore structure and rigidity in the annealed sample, therefore, the G' values should be higher.Consequently the increased structure will lower the materials ability to dampen out the energyplaced into it through the sinusoidal strain. This would be indicated by lower values of Tan Delta.This behavior can be seen in figures 6 & 7. The sample designated A was heat treated andexhibits higher G' values and lower Tan Delta values than the sample not annealed.

CONCLUSIONS

Dynamic mechanical testing is able to give insights into the microstructure and macrostructureof the liquid crystal polymers selected for this program. The annealing phenomenon wasdetected in the granlar resin. The testing done thus far is very preliminary, clearly more workneeds to be conducted to characterize the rheology of these highlycomplex polymers.

213

_ _ _ __ _ _ _ e

In

Cw 4

\ \%

~ \ \\

0I

% %

0) Uaw/UxD] .9 CV) lawa/uAp) .9

214

FIII1 t I I I

CLICLC

in 4-0Cu

(

Lz~10

E- 3- 0 LL

a.-I

a. it

IL

w.Iq< I

Ii 9ccq.

oM:cc4

a-

0)EslI).9 V oo-P

L21

II

LM1u9

U)

C.)

U-p

o 4lC.)

a-t

z4cI

4 9

-21

TASK #9: MATERIAL PROPERTY DATABASE

Thomas A. ElkinsOL-AC PL/RKAD

Edwards AFB, CA 93523-5000(805) 275-5303, DSN 525-5303

E-mail : ELKINSTOPL-EDWARDS.AF.MIL

ABSTRACT

Presumably the researchers working on discovering the fundamental microscopic and macroscopic prop-erties of these liquid crystal polymers would have a place in which to store their data so that it would be

readily available to the other researchers and part designers, and, indeed, there will be one.The purpose of this task is to develop a database management system (DBMS) capable of storing and

manipulating the material properties that will be discovered as the research into liquid crystal polymerscontinues.

DATABASE DEVELOPMENT

Select conceptual data representation

Develop logical data representation

Develop data manipulation language €* Develop data definition language

Develop syntax checking and parsing algorithms ®

Develop pseudocode for database operations 0

Refine and modularize pseudocode

Convert pseudocode to real code

Compile & test code

Debug . 0

Develop physical data representation

Develop advanced user-interface

Develop plotting capability

Develop report-generation capability

Develop data exchange capability

Write docunientati,,i

Publish

0

217

Sub-task 1 SELECT CONCEPTUAL DATA REPRESENTATIONStatus : Done

Originally, the database management system (dbms) was supposed to contain only polymeric materialproperties. Such a system can, and was, developed using a standard record-based system; however, thescope of the project was expanded to handle metals, ceramics, and composites. A conventional record-basedsystem is not feasible because of the inconsistency of the data fields to be represented (a good example wouldbe glass transition temperature which is a valid piece of data for ceramics and polymers, but not metals)and the variability of data (some material properties are highly temperature dependent, others are timedependent...). To represent those dependencies and inconsistencies with a record-based system would wastememory, slow data processing, introduce redundant data, and possibly violate data integrity.

A more feasible approach would be to develop a relational system. Relational database systems arehighly flexible, but very complex for both the user and the programmer. A record-based system provides theuser with a template, or form, into which the user can enter data, browse, search, or manipulate contents.The relational system allows the user to think of his/her data as being in tables or sets and provides the userwith tools with which he/she can manipulate the relations to create new relations with the desired data.

The following is an example of using a record-based system vs. a relational system for an employeedatabase. The data to be stored are the employee I.D. number, full name, home phone, date of birth, inwhat group he/she works, and the projects on which he/she is currently working.

Record-based systemLDr. tNumbz Last Name Bid Nam Home Phone D.O.B. 2Prject123466769 Doe John 123-4567 500321 ABC T15123456789 Doe John 123-4567 500321 ABC P2123456789 Doe John 123-4567 500321 ABC X8987654321 Johnson Mike 555-1212 620513 ABZ P2987654321 Johnson Mike 555-1212 620513 ABZ Q4393939393 Phillips Lisa 921-3484 451005 AB234970923 Fisbin George 636-2342 540612 ABZ

Notice that all of the required information is stored, but there are data redundancies (name, date ofbirth, phone, I.D., and group do not change because the project changes), and potential problems for dataintegrity (if John Doe changes his phone number, every record with his old phone number must be changed).Also, this database system is limited as to the information that can be extracted (Who is Mike Johnson'ssupervisor? Who is the project manager for P2? What are the phone numbers for everyone in ABZ?). Fieldscan be added to the database to reflect the desired information, but more problems are introduced, i.e. noteveryone will be a supervisor or project manager (introducing empty cells), and although everyone has asupervisor, there is not a different supervisor for every employee (introducing more redundancy)

Relational systemEmployee-dataID. Number Last Name First Name Home Phone D.O.B. Grou!123456789 Doe John 123-4567 500321 ABC987654321 Johnson Mike 555-1212 620513 ABZ393939393 Phillips Lisa 921-3484 451005 A B234970923 Fisbin George 636-2343 540612 ABZ

Projects Siwervis,,r

T5 123456789 t% BZ 234!70923P2 123456789 AB 393939393X8 123456789 ABC 123456789P2 987654321Q4 987654321

This schema also contains all of the required information, but notice that the redundancy has beenminimised. Also the relation Supervisors was created using as little information as is needed to representthat information; similarly, a relation of Project Managers could be created. To extract information, the

218

user performs operations on relations similar to operations performed on sets. For example, if we wanted tofind Mike Johnson's supervisor we could perform operations on the Employee-data and Supervisors relationsto extract the required information. Similarly, if we wanted to get a phone listing of everybody in the AB

division (including the branches), the user would create a new relation describing the relationships betweenbranches and divisions and then use that relation with the Employee-data relation to retrieve the necessaryinformation.

The relational system allows for tremendous flexibility which makes the job very difficult for the pro-graminer who no longer knows what data will be stored or how the data will be used. The programmerprovides the tools to use on relations, and the user is left with the responsibility of using the tools to defineand manipulate relations to obtain the desired information.

Sub-task 2: DEVELOP LOGICAL DATA REPRESENTATION

Status : Done

The logical data representation is the way the data will look to the computer and, as will be seen, ismuch different than the conceptual data representation. The conceptual representation of a relational systemis of a series of tables of data (as might appear in a book). The logical data representation consists of thedata structLres the software will use to store and manipulate the user's data and serves as the basis uponwhich a physical data representation will be developed.

The following is a list of knowns and unknowns of an implementation of a relational database system.KNOWN UNKNOWN

A database has relations Number of relations in a databaseRelations have... Number of attributes in a relation

"a name Number of tuples in a relation"a creator Number of domains in a database"a set of attributes Relationships between relations"a key attribute Values in cells"a set of tuples Value types of cellssecurity restrictions Which cell of a tuple is used for sortingusers

Attributes have...a name

"a set of cellsTuples have...

"a set of cells"a key cell

Cells have valuesValues are from a specific domainEach cell is associated with an attribute

Because of the unknowns, a dynamic memory allocation scheme will be used. The maximum number ofrelations, attributes, tuples, and cells will be determined by the computer's memory and operating system,and backing store if the operating system employs a virtual memory management system (virtual memoryoperating systems use peripheral storage devices as secondary memory storage, and are t' as not limited toon-board memory).

The programming lnngiiage that will he used on thi! prji.r-r 0 C'. whirh c-nprates vwrv fnt., port.nhlecode which will allow the code to be converted (ported) to other machines with 'ry lit ti. ifnnv. ,iogification.The C data structures that will be used on this project arc as f,,llvwsýstruct DOMAIN I

char name[20]; /* domain name *unsigned type : 4; /* data type for domain */float format; /* data format (like printf) */int instances; /* times domain is used*/struct DOMAIN *next; /0 pointer to next domain */struct DOMAIN *prev; /* pointer to previous domain C/

219

struct FIELD {char name[20]; /* name of the field */etruct DOMAIN *domain; /0 domain of field */utruct FIELD *next; /* pointer to next field */struct FIELD *prev; /* pointer to previous field */I;

typedef struct FIELD *FIELDPTR;

struct CELL {FIELDPTK field; /* pointer to field info */union {

int *i; /* pointer to an integer value */long *1; /* pointer to a long integer */float or; /* pointer to a real number 0/double 0d; /* pointer to a double real 0/char 0c; /* pointer to a string $/float 08; /I pointer to scientific notation '/

} value;struct CELL *next; /* pointer to next cell '/struct CELL *prev; /* pointer to previous cell '/struct CELL *up; /* pointer to cell above */struct CELL *down; /* pointer to cell below '/

typedef struct CELL *CELLPTR;

struct TUPLE {CELLPTR fivetrcell; /* pointer to the first cell l/CELLPTR key-cell; /* pointer to the key cell */CELLPTR last-cell; /* pointer to the last cell */struct TUPLE *next; /* pointer to next tuple */struct TUPLE *prev; /* pointer to previous tuple '1

typedef struct TUPLE *TUPLEPTR;

struct RELATION {char name[20]; /* name of the relation '/char owner(15]; /e owner's username */char file[80]; /* file specification of data '/unsigned pread 1; /* World can read & use relation '/unsigned p-add 1; /* World can add tuples */unsigned p-modify : 1; /* World can modify cell contents '/unsigned p-del 1; /* World can delete tuples */unsigned ascend 1; /* direction for sorting key '/unsigned altered : 1; / t to I if data Pnr altered */char user[15]; name of last user *ichar date(6]; /* date relation was last used */int fields; /* number of fields */int tuples; /* number of tuples */struct RELATION *next; /* pointer to next relation '/struct RELATION *prey; /* pointer to previous relation '/FIELDPTR first-field; /* pointer to first field '/FIELDPTR key-field; /* pointer to key field *I

220

FIELDPTR last-field; /* pointer to last field ,/

TUPLEPT& firat-tupl"; /* pointer to first tuple */

TUPLEPTR current.tuple; /* pointer to current tuple */

TUPLEPTR last-tuple; /* pointer to last tuple */

NOTE: the protection flags will be used for security purposes. The user's username is first compared

to the owner field; if there is a match the protection flags are ignored (the owner can do whatever he/she

wants). If the user is not the owner the protection flags affect the following operations: "seeing" the relation

(READ), adding tuples (ADD), changing the contents of a tuple (MODIFY), deleting tuples (DEL).

Sub-task 3: DEVELOP DATA DEFINITION/MANIPULATION LANGUAGEStatus : Done

The data definition/manipulation ' iguage is the set of commands available to the end user for creating

and modifying relations. The language defines how relations, and data within the relations, can be manipu-lated and modified. The selected language operators and syntax are listed below in the BNF (Baekus-Naur

Form) notation.defined-as, or, [] : 0 or 1, {} 0 or more

DEFINITIONS"< letter > ::A ABICIDIEIF}... X}YIZaIbIcId}... wjx~ylz

"< digit > ::= 0111213141516171819"< symbol >::=!IOI#ISI%j A I&1 * I(I)1-I - I + I = I'l [I{!]1}I; I: I'I"I \ I < I > 1, 1.1/1?"< integer >::= [+I-] < digit > {< digit >1"< real >::=< integer > .{< digit >}[Ele < integer >]"< name >::=< letter > {< letter > I < digit > 1-1 - I#j%}< domain >::=< name >"< domaintype >::= CHARACTER I INTEGER I LONG I REAL I DOUBLE I SCIENTIFIC"< number >::=< integer > I < real >

"< value >::=< number > I < name > I{< letter> I < digit > I < symbol >)"< relation >::=< name >"<field >::=< name >< condition >::== I- I < I<= I> I>= I?> ~?< attribute > ::= NAME I KEY [ SORT I PROTECTION< fieldlist >::=< field > [,< fieldlist >1< fieldef >::=< name > : < domain > [, < fieldef >1

< predicate >::=< field >< condition >< value >Commands

CREATE < name > WITH (< fieldef >) [USING < field > UPI DOWN]DELETE RELATION < relation >USE < relation >JOIN < relation > TO < relation > WHERE (< field >=< field > {, < field >=< field >)UNION <" relation > AND < relation >

INTERSECT < relation > AND < relation > ONTO < name >DIFFERENCE < relation > AND < relation > ONTO < name >COPY < relation > ONTO < name - (WHERE .- prodircnf .{ AND I OR .• predircn.. ý'}1SAVECH1ANG;E RELATION - attributc -. To l,, I. ,a,,,t If ) ,,

SIIOW j< relation >[i,ISTDEFINE < name > AS < domaintype > numberUNDEFINE < domain >ADD (- fieldef >) [TO < relation >1REMOVE (< fieldlist >) [FROM < relation >]CHANGE FIELD (< field > TO < name > {, < field > TO < name >}) (IN < relation >1PROJECT (< fieldlist >) ONTO < name > [WHERE < predicate > { AND I OR < predicate >}]

2221I I

COUNT < field >SUM < field >AN? H ItAGE < field >MAX < field >MIN < field >MULTIPLY < field > BY < value > [WHERE < predicate >]INCREASE <field > BY < value > [WHERE < predicate >1DECREASE < field > BY < value > [WHERE < predicate >1DIVIDE < field > BY < value > [WHERE < predicate >]SET < field > TO < value > [WHERE < predicate >1FIRSTLASTNEXTPREVIOUSINSERTEDITDISPLAYDELETE TUPLE [WHERE < predicate > { AND I OR < predicate >}]SEARCH FOR < predicate > { AND I OR < predicate >}FIND

Please see appendix 1 for a description of each command and how it is used.

Sub-task 4: DEVELOP SYNTAX CHECKING AND PARSING ALGORITHMSStatus : Done

This step involves separating words and symbols from a line entered at the keyboard and comparingthem to a syntax template to determine if the user entered the correct command sequence. Naturally, thisstep requires that the syntax for the language be established. Once parsing and syntax checking is complete,the semantics of the command line must be checked (i.e. the user entered an existing relation/field name)

Sub-task 5: DEVELOP PSEUDOCODE FOR DATABASE OPERATIONSStatus : Completed for current DDL/DML

Using the logical data representation and the database language, describe how the commands willfunction. For example, the command JOIN might be described as follows:

Function JOINcreate new relationadd fields from first relationadd fields from second relation

make cross-product of tuples from both relationsinsert tuples into new relationdelete tuples that do not meet requirementsadd relation to database

Endfunction

This must be done for all commands.

Sub-task 6: REFINE AND MODULARIZE PSEUDOCODEStatus : Completed for current D ,I)r MI•,

The pseudocode routines are further refined, coming closer to actual codc. Frequently recurring state-ments are tagged for possible development as separate modules which could be used by multiple routines(such as the "add fields" lines listed above - there is an add field command, so the same routine might beused). Also, the logic of the routines is checked every time a change is made. Using the example from sub-task 5, the "make cross-product" statement definitely needs to be refined because there is no cross-productroutine in C, and certainly not one for the data structures used in this code.

222

Sub-task 7: CONVERT PSEUDOCODE TO REAL CODEStatus : Completed for current DDL/DML

Now comes the hard part, taking the conceptual routines and writing real code that will actually work.

Sub-tasks 8 & 9: COMPILE & TEST CODE / DEBUGStatus : In progress for current DDL/DML

This is the step where the code is converted to an executable and the routines tested. Compiling willreveal any syntax errors from sub-task 7, linking will reveal any semantic errors, and testing will reveallogical (run-time) errors. Debugging involves returning to sub-task 4 or 5, finding where the logic fails, andcorrecting it. A test case is developed to test the code and the various DDL/DML routines. Please seeappendix I for the actual test file and appendix 2 for the output from using the test file. This is the timefor any interested users to give their ideas for features/capabilities to be included in the code. (hint, hint!)

Sub-task 10: DEVELOP PHYSICAL DATA REPRESENTATIONStatus : not started yet

Once the logic has been proven the logical data representation is finished and it is time to move on tothe next level - the physical data representation. The date.base is useless if the user cannot store data foruse later. The physical data representation is the method b3 which the code transfers data from memory tobacking store (hard disks, floppy disks, etc). Some of the methods are sequential access, random access, andindexed, each of which have applicability in database management. Sequential access, as the name indicates,looks at each record in the order it was written to disk - this would not be good for searching, but sincethe logical data representation calls for keeping the data in memory, sequential access is the cheapest way tostore the tuples of a relation. Doing this will require a separate file for each relation; however, it also meansthat the entire database will not occupy memory, only the current relation and a few others (for doing ajoin or union). The relation headers will be in a separate file (accessed when a user selects a database) andcan use any method, but since the headers are small they will probably be stored sequentially and kept inmemory. Sub-task 11: DEVELOP ADVANCED USER-INTERFACE

Status : not started yet

As is, the software will process user requests through the command line interface and syntax parser;however, this means that the user is required to learn the database language. Currently, it is possible tostore commands in a file which is then read by the database syntax parser and executed (this is how thedatabase is being tested). To make the software more user-friendly, an interface must be developed thatwill make the processing and querying functions easier for the user and be able to translate them into theequivalent language commands. Graphics-oriented interfaces are populer, but require a tremendous amountof development and testing time and must be developed for each platform (not all computers use the samegraphics). It would not be uncommon for a graphics user interface (GUI) to be larger than the code forwhich it is used. A majority of users will be using this software on the VAX systems, so the primary focuswill be a user-friendly interface suited for VAX environments. Other interfaces may include an X-windows

driver for using the code on remote workstations. Of course if the user does not have access to a graphicsterminal, the command line interface is always available.

Additinnally, extensions to the current database langling' nnd m..r.,,ifirRtionf no the 1pnrqer would nl-low the user to write "scripts," simple programs which the dhmis "•-vild ex.e,-r I-, prtrf,,ti ih,, datn en-try/manipulation/extraction tasks for the user autornaticfllv.

Sub-task 12: DEVELOP PLOTTING CAPABILITYStatus : not started yet

A plotting capability is not normally provided with generic relational database systems; however, consid-ering the primary purpose for developing this system is to compare material capabilities and applicabilities,it would be very helpful to visualize how one material changes with the environment or how several materialscompare with one another. Again, the graphics will be different for different platforms, so a plotting modulewill be another large program. It is unknown at this time whether to add a plot command with the supplied

223

database language or tie it to the user-interface. Also unknown are the types of plots and user-controllableattributes. Inpnt from all interested users is highly encouraged.

Sub-task 13: DEVELOP REPORT-GENERATION CAPABILITYStatus : not started yet

A database is not completely finished if one cannot get data out of it. A researcher may need to showtables of data in his/her report, or a program manager may need to show why he/she came to a particularconclusion; therefore, the capability to pull data from the database and put it into a format that can beincorporated into text documents is important. A standard ASCII output file will, of course, be included,as well as a W file. User-supplied formats will be considered.

Sub-task 14: DEVELOP DATA EXCHANGE CAPABILITYStatus : not started yet

Data can come from a variety of sources, so the capability to take data from external sources must beprovided. The process is to write data translators which read data from the format of the source and writedata out in the format required by the database software. Since the programmer cannot know every possiblesource, there must be ways for user-supplied programs to get data into the database. Similarly, data maybe needed in other codes (finite element codes need material property data) and must be translated. Onepossibility is to deve' )p a "universal" file format by which data can be exchanged. Definite translators willbe included for the I-DEAS family of analysis codes and also for Lotus 1-2-3 data files (.PRN). Interestedusers are encouraged to provide data formats with which they are familiar.

Sub-task 15: WRITE DOCUMENTATIONStatus : not started yet

This task involves writing a detailed user's manual.

Sub-task 16: PUBLISHStatus : not started yet

224

APPENDIX 1 : Sample Input Deck

The followi-Ig pages list the input file used to test the parser, syntax checker, semantics checker, andcommand processors. The input file also gives a brief description of the command being tested and sometiotes about capabilities. The file creates and manipulates a database of data from the periodic chart.

The code was developed on a VAX system running VMS 5.4, but was d--o ported to a Tektronix XD88workstation running Unix V5 with no changes to the code. The input file was generated on the VAX and

FTP'd to the Tektronix, also with no changes. File redirection was used on both platforms. For the Unixsystem. the command was

mpdb < test.dat > test.outFor the VAX, the commands were

S ASSIGN TKST.DAT SYS$INPUT$ ASSIGN TEST.OUT SYS$OUTPUT$ RUN PPOB$ DEASSIGN STSSOUTPUT

2

225

0 Sample input file to test the commends for the relational databaset management system (ROBSS).# NOTE: conventions used in this file are as follows...4 Command syntax descriptions are proceeded by a line of asterisks '

# Items in brackets '(I' are optional items which may be used once por command.I Items in braces ()' are optional items which may be used many times.# items which are in all capital letters are required keywords.* Items delimited by a vertical bar 'I' separate the list of valid keywords# which may be used.I Punctuation marks (other than those listed above) ace required where shown.

I DEFINS domain name AS domain type field widthl.decimal places)# Values in a field must belong to a speciTic domain.# Domain type may be one of the following:I CHARACTER - for textual dataI INTEGER - for single precision integer data# LONG - for double precision integer datat REAL - for single precision floating-point dataI DOUBLE - for double precision floating-point dataI SCIENTIFIC - for scientific notation formatdefine atomic name as character 20define junk sa scientific 13.6define atomic symbol as character 3define temperture as real 7.3S

#LIST - Brief list of the contents of the database.listS

* DELETE DOMAIN domain name or UNDEFINE domain name0 Delete unused or unwanted domains.undefin* junklist

I CREATS relation name WITH (field name : domain name flield name s domain name)) (USING field name UPIDOW II Relations are c;llections of det; organized by--fields (columns) "o0 and tuples (rows). The user defines how the data in a relation is0 organised when it Is created.create elements with (

name a atomic name,symbol : atomic symbol,boil I temperature,melt temperature

using symbol up-.... orts the data in this relation by symbol in ascending order.

*If the optional USING clause is not used, the default is to sort the# data by the first field in descending order.I Note that BOIL and MELT are from the same domain. Multiple fields* may be drawn from the same domain, but fields may not have multipleI domains.list

* SHOW (relation name]* Show detailed Tnformation about a relation. If relation nane is not* specified, the current relation is displayed (if assigneo).showshow elements

0 USE relation name* Koke the rolltion current.* Some operations are performed only on the current relatiin.use elements

# INSERT (data)I The INSERT command will prompt you for input for each field:* however, if you know the order of the fields, the data can be* placed on the command line (unless you have character dataI that say contain spaces).# Note also that the input can be placed in tabular form.* (useful for Incorporotirg data from external files)insect Hydrogen H 20.268 14.025insert Helium He 4.215 .95insert Lithium Li 1615 453.7insert Beryllium Be 2743 1560insert Soron a 4275 2300insert Carbon C 4470 4100insert Nitrogen N 77.35 63.14insert Oxygen 0 90.10 50.35insert Fluorine F 84.95 53.48insert meon Ne 27.096 24.553 226

0# There should be 10 tuples in the ELEMNI~TS relation.0 There should also be an asterisk 1- next to ELEMZNTS, indicating# it is the current relation.list0 .............. s....e..............................................* PROJECT ( field (, field) ) ONTO relation name (WHERE field condition value (AN0leS field condition value# copies columns of date from the currant r~lation ta a new relation.project (name) onto namesI No qualifier was used, a* the NAME field of all tuples wes copied.listus* names

# ........................................

* FIRST N- oves the current tuple pointer to the first tuple in the current relation* NEXT -- move$ the current tuple pointer to the next tuple in the current relation# DISPLAY -- displays the contents of the current tuple.# Note how multiple coamands may be placed on the aome line ...first displaynext displayIuse elementaproject (boil~name) onto boiling pta0 Two fields of all tuples are copied.listshow boiling pts0project (nsine,Symbol) onto hot ones where melt >1000 Two fields are copied, but oily for tuples that mEest theI stated requirement.*The conditionals that are accepted are as follows:

* - Equals 1. Does not Equal# Less Than m LeeSS Than or Equal# Greater Than Greater Than or Equal# ? Contains 17 Does Not Containlistshow hot-ones0project (symbol,melt,name) onto junk where

name I? andboil (-1615

1 Three fields this time, but the requirements are more rigid.# The qualifier "name 17 e' means select entries in the fieldp NMAE that do not contain the letter 'e' flike "Lithium")

s1how junk

$ DELETE RELATION relation name# Self explanatory.delete relation namesdelete relation boiling ptsdelete relation hot-onesdolete relation junk0................................ * ...... a......eaa~e.eeeeee~eeeeeeee

I COPYf relation ONTO new relation (WERER field condition value (ANDIOR field condition value))# Copies specified tuple; with all fielde.# (unlike PROJECT which copies certain fields)copy elements onto cool ones where boil t- 100listuse cool onesshowfirst displaynext displaynext display

0 create a new relation with different data

define atomic-nuaber as integer 3define atomic weight as real 10.5create atomic data with(

number atomic_number,symbol atomic symbol,weight atomic weight

listshow atomic s.atsuse atomic- datashow04 Insert the new data.4insert 1 Ne 1.0079insert 2 He 4.0026insert 3 Li 6.941 227

insert 4 s. 9.01218insert 5 3 10.81insert 6 C 12.011insert 7 N 14.0067insert 8 0 15.9994insert 9 F 18.998403insert 10 me 20.179insert 11 ga 22.98977* Since no gotting directive was given, the default is the firstI field Listed in the CREATZ parameter list. Also the default* sorting direction is in descending order, so the first tuple* of this now relation should be the one with the highest number.* (Sodium - 811)first display

JOiN relationl TO relation2 WHRRE ( fieldl condition field2 (, fields condition fieldn)I Joins two relations so that the data from both is incorporatedI into one relation. The acceptable combinations are determinedI by the condition(a) specified.join atomic data to elements where (symbol - symbol)

# The data is grouped together using the fact that they share SYMBOLI information. Notice, though, that ATOMIC DATA contains a record* that does not match any in ELBMEUTS. ThaT record is skipped.

e 2LeMENTS should now have the same number of tuples (10),I but two more fields (total 6 fields).

listuse elementsshow

I* Display some of the tuples to make sure the data is correct.8

first displaynext displaynext display6*...aeaoea~eaaaeteae..eeaoaa...aeeoa..see..aeso...aaoeaaeao~t....e.*flea...

# CHANGE FZILD (old field name TO now field name (, old TO nowI Rename the fields-to sojething moroe--aesoablo (fields are renamed* to avoid the possibility of duplicate names while joining)change field (

elemente name to name,elemonts-symbol to symbol,elements boil to boil,elementsc datt to melt,atomic mdltenumber to number,atomic data weight to weight

8# We don't need ATOMtC DATA any more, so delete it.Idelete relation atomic datalist

# Make a noa relation with the same fields, but now data.S

create table withname : atomic name,symbol : stomic symbol,boil temperature,melt : temperature,number i atomic number,weight : atomic weLght

using melt downI tqt

u-.@ table

showinsekt Sodium Ma 1156 371.0 II 22.96977insert Magnesium Mg 1363 922 12 24.305insect Aluminum Al 2793 933.25 13 26.96154ijiert Silicon Si 3540 1685 14 28.0855insert Phosphorus P 550 317.3 15 30.97376insert Sulfur $ 717.75 388.36 16 32.06insert Chlorine Cl 2V9.1 172.16 17 35.453insert Argon Ar 87.3 83.81 18 39.9485

first displaynext displaynext display0

* .................* UNION relationl AND relation2* Appends the contents of relation2 onto relstionl. The twoI relations must be compatible (some I of fields and matching domains0 for each field). 228

list # beforeunion elements and table# of course we could hiv. also inserted the data into EL!'EMENTS4 directly, but this illustrates how relations can be appended.list 0 aftershow elementsUse elementsfirst display

next displaynqst di-.playnext display0

I ADD ( fieldl domain name (. fieldn :domain name) ( TO relation)0 Adds fields to an exiliting relation. In this example,# we add a new field to the current relation to etore the physicalI state of the element at 305 degrees Kelvin.define state as character 6listadd (state :state)I No qualifier since we are adding it to the current relation.# Note that a field may have the same name as a domain.show

0 511T field name TO value (WHERE field condition value (ANDOIR field condition value)!* Sets all or some of the field values to a given value.set state to solid# Moat of the eloements are solid, so we not all of the STATE values0 to 'solid, for now.0first display

0 Next. we set the STATE value to 'liquid, for all elements whose0 melting point is below the specified temperature.

set state to liquid where melt 4 305first display

0 Do the same thing for gaseous elements.I (boiling point <temperature)0set state to gas where boil < 305first display

net displaynext displaynext displaynext displaynext ipanext displaynext displaynext displaynext displaynext displaynext displaynext displaynext displaynext displaynext displaynext displaynext display

..................................... aaaaaeeaeCbaaaeeesa

# DELETE TUPLE (WHERE field condition value f(ANDIOR field condition value))* Deletes either the current tuple (if no qualifier added), or0 specified tuples.delete tuple where s tats - solid and number ( 70 This should have deleted Lithium, SerylLium, Baron, and Carbon.first displaynext displaynext displaynext displaynext displaynext displaynext displaynext displaynext displaynext displaynext displaynext displaynext displaynext displaynext display

: ............................................a..........0 REMOVE Ifieldl (, fisldni ) (FROM relation -name)* or DELETE FIELD (fieldi 1, fio'

4n) ) (FROM4 relation na 2b

0 Deletes a field and all associated values from a relation.04 Let's remove the STATE field so that ELEMENTS and TABLE will* be compatible.remove (state) from elements

show elements

I Since STATE has no instances, we could delete it now.I

* Add now data to TABLE that will not be in ELEMENTS.use tableinsert Potassium K 1032 336.35 19 39.0943insert Calcium Ca 1757 1112 20 40.08insert Scandium Sc 3104 1812 21 44.9559insert Titanium Ti 3562 1943 22 47.9S

I INTERSECT relationi AND relation2 ONTO now relationI Now, lots find the Intersection of ELEMENT! and TABLE.# ithose elements that occur in both ELEMENTS and TABLE)intersect elements and table onto intersectlistuse Intersectfirst displaynest displaynext displaynext displaynext displaynext displaynext displaynext display

# DIFFERENCE relationl AND relation2 ONTO new relationI Now let's find the difference of the two relations.I (elements occuring in one but not the other)difference elements and table onto differenceslistshow differencesuse differencesfirst displaynext displaynext displaynext displaynext displaynext displaynext displaynext displaynext displaynext display0***.*4*C.ea.*eCe*eaeoeoea.aa*Cetoeaeao~eta..estoeoesaaaoeee.***..e~a***h..*.

# SEARCH FOR field condition value (ANDIOR field condition value)I Searches for tuples in the current relation which satisfy the* given conditions.search for symbol I N or number o- 20displayS

•eee~e~eee eca e. eae .eaa eeee.......... ..eaeca C.....eeea Coe*

# FIND -- searches for the next tuple using the conditions from theI last SEARCE command.find displayfind displayfind displayfind displayfind display

0#*~ .t......e......aaaa......... ..t.......*.S I have included a few statistical operators. Their function is0 straightforward, and all use the same format.0 COUNT field - returns the number of items in the field.* SUM field - returns the sun of the items in the field.* MAX field - returns the maximum value in the field.0 MIN field - returns the minimum value in the field.0 AVERAGE field - returns the average value in the field.count meltsum weightmax boilmax symbolmin namemin weightaverage weightaverage symbol

Ij I have also included math operators which can take qualifiers

230

6 INCREASE field BY value (WHERE field condition value (ANDIOR field condition value))

I DECREASE field BY value [WHERE field condition value (ANDIOR field condition value]l

I MULTIPLY field BY value [WHERE field condition value [AIDJOR field condition value))

DIV'IDE field By value 1WHERE field condition value ([HOAOR field condition valuo))

increase boil by 1000 where number ý 15decrease melt by 1000 where boil < 300multiply weight by 100 where name I? a # name does not contain an I'tdivide number by 2 where symbol ) M 6 symbol alphabetically follows 'N'

0 1 number I old j boil I old I name I old symbol I oldI Element I ) 15 boil 4 300 I melt t? a weight ). K I number)

#--- ------------------------------------------------------------------------- Ifirst display #CALCIUM I yes 1 1757 1 no 1 1112 1 yes 1 40.06 I no I 30next display #FLUORINE no 1 84.951 yes 1 53.481 no 1 18.99041 no 1 9 1

next display #HYDROGEN I no 1 20.261 yes 1 14.021 no 1 1.0079 1 no Inext display 1HELIU I no 1 4.2151 yes 1 0.95 1 no 1 4.0026 1 no 1 2 1next display IPOTASSIUMI yes 1 1032 1 no 1 336.31 yea 1 39.09831 no 1 19next display #NITROGEN no 1 77.351 yes 1 63.141 no I 14.00671 yes 1 7 1next display @NEON no 1 27.09j yes 1 24.551 no 1 20.179 1 yes 1 10 1

next display #OXYGEN no 1 90.181 yes 1 50.351 no 1 15.99941 yes 1 0 Inext display #SCANDIUM yes 1 3104 I no 1612 yes 44.9559g yes 1 21 1

next display OTITANIUM yes 3562 I no 1943 yes 47.90 1 yes 22listquit #deletes all domains and relations and exits to the operating system

231

APP3NDIX 2 : Output From Sample Input Deck

0232

pc0V

4-

c- c

* c Ir

0 42 4) 0, V*4 C 2

*4 m MU~ 14 c cC 41A

4)~~~~6 4). 4 3,)~mC

.9. 4!0 VUUU 4)1 UUU '14). -CA .*1 ic - : -C4 .44 CC a *% SIc

4)0 04 4 a4

z A AS a Ua4)~ IL I w)p V4~9 4 .

)C4) 4)9 4 0) ., e4

.. d~ v) L I 4) 20%2 4 E.

44. 0* 0J-4) a) 0 4.UO 1)9-L 06 0

a *.LI -9. . 49 1

64 .4 4, V . p .. 4 .40

4 4 .. -.ar j.9 00

16 *,A -, I4 *9 4 .. "4 ... . . . . . . . . . . . . . . . 0 . . .

ol VI . * 91 -Ci C 2 2 ! 4 C C Z C C UC C4C 0)L4 4 4 9 4 U 9 9 4 4 4 S P 6 C A 1-94 495 49zIrx2Iu .I II Ic 2

uL lu V w u .. - .. c.. u. .

p 0 -233C

o 4?

o 4:

0 c 0 4$c C 0 w6

a$ 4$iý x

V$ . 4414 4$ 1

CL$ 4$ Cs C$ *-C

1.0 *1 Iv 4t 0 4$ *C

IV W. -C V- 45 4. L c X

*~~~~C 4.4 t 4~0V* 0$ *$ ,4 C$ 61, 0 ~

2P g4 4$ p C c '0 0O L L U 1$ L a8~~p- 0$ L L4 V$ C 4$ 0 a

4$ C.i CA$ CS Ci'-J 1* .. U'o'

0 L 4$ .4 4 4$ V C$ ~ r -C r - C .*4~~.4.4o'~ C- IE 2$ C $ '. - . . . . C

C..4 4e 1", 4$ o 0 4$ *1 C-00 4 46, 4$ 4$ ag u~ -,

ItIL 3 6.4 a$ Id 4$ 0 ~ 1

&4k U ac 20 A$ 4' LC L* L L L

al.. 4$ t4 .$ , 4$ tV.. c C*-p CS0 4 0 -W 4$ I6 4$ V z$ OL . .. .. . t. '

x4 4$ z* 0' 4$ c' 4$ c C r r' ..- C C!'' 'C

22VvIV ' :. v 1 V a'U1 4$D 04. 13 'a 10 v6 V$ v 'D '00v v

a 0S * a$ V.. aj ftP Vol

*9.4 -a w4 ..uu-It u Z00. -- 41$ u su 4$ u 410 w u %i si u u Q u

*t 4 '0 16 ' 4 . $ V U t

* .46 96.2 C C' 4$c4 CX C2346

Icr

0 &*.c2

.41.

c C

Cý ~ ~ ~ .la.-C

;11 ~ ~ ~ I .44 . . .2

t 4. 6 C 0C.44 I. c . 9.

400

f:- : - * 4U

4:c. 3. Gv , . T

LC '.4 -a 4 24: .., 1.8 1 .1 ZCCL V

V . 4 : . 4 0 *rV X ,C

UUU 4..2 90 c. 4 2 .. .. .. .....

049 L 4 -. -4 *4.. '

m a-' 44l. lb 0. . 4. 44 2 W4.4, c is01 t . *.* "I .. * -. 0. , a9CUVLI I. L C ..0

-4 40 - 4"

. 449 VE QC .n a, a ou 3. Vv v . t

10 m 0 V.9 04 LC an OC'4- .0 a9 c ýq L 0v 44.

'A LL C 222 -. 1

4'.4~2- 4144 -. ...V O LU V V@99' 649 4' A4 r. 0044 aC44 a -j u

0 '0

19 4 .01 49 %I 0 av44 * v 4440 1 00' 7VvV v vV2 n. 0' 0 0 V v'

10 1 1 1 4 aL 0 M4 *C0' M 4 L L.1 5C MI aI S. ' f09 . .j. 45 a C0ý 0.0 asI37'1: . ... sI 6..a3 x7.j. j 411 PC94 0 a4! ,p rwx jj K 10aC.*C0 0C ,v ,k 00 o t 0 C. . a m010L1

m~. 0', Cf -V k jwu1 naZuut u j"SC lýa 0 MLC L. L - . j uk u

* ~ * *4.2 C2 *02354

.4w 0 nC 0 U

C C L0 a. '' c'

SM.4 00 #1 '10 t-

a' cC U Lc C.C 6

0001 .4~anU. - .

aCI- L C a'C ~ C .

o~ j .%DC-''0CL-e .-. CL.. C . . .. L.L aJ

2 22 2 6. 7 2 C 24*46 0CM +1 4- V V a'-

-"1 0 0* 0 0* *' L

U, ~ -4 'a,! V .

4 0 -4 4, 4 a' c 6.,. -,aU of w 44 C,

.C .. 6.L. .. .. .. .. .. LL ..--v -. 00 0.6 ;00;

U- U-U .4 PEO v a-L~ ~ ~ ~ c n 00 m' t

.4C 4 * 6 SC 0*C. L ~a 0.. IV 41. c , *'' v , *'4'V4.Cc

CC C)6. m.*C0I#P L4.C, C .*. orCOc C vVc0 U- 4. 41 a 0--U w o10-4' V 41 0 4 0e -.

.0 c0 -C 9e. 0.. J 3r c ' 12 . n

aI a - ;c LLmCI- b. '! *0. -,

tL +1. p. v - C* 6

£6 .4 -aj~w cn el, -M C.48- *.4 C 6. ..'.. .. .. .. .. .. .. C L*~ ~8 .. .. ..

.. .. .. 6. C.. .. ' 'a 0 'a r 0 U ' C C ' CWIJCJ~~. .4 04 .08 . '. 0 ..0. 0' .0

3t~ C.£ 8U * I c IC.U FUc c'0 8

0.41 14 C0 40*- TO -1£ ).0.C 0'4 2h 4-It-e a C, rc C..4 . 4 - P ' . C

u 1wur , L- w -a~ j u Q0 a' v u Qw 0 C; w.C a0*5z t L . wI-

-~~I L ~.go

* 0 7 0 IC-Fl'0236

Ic0

.4

0u

4)4

41'* C L 11

X4 w 4

ot X

c Q1' 445 1

4-.L* 41'

c IL

4: 1 UI41 (,14 U'445 454 .44

I. 10 6~5. a 1 Ll

4,-4 L '.44- q4 a 4'

4L' u o U c cIL 4 C. r 0.4C

L.. 4 N5 ... 14 *1' I&W1

.14 .4I . 4. *u

Sr 4a. 1. 0 am " mm I m m L .

4LU . S .- 4

u U -- 04ý juuj ue 4.) a o m 0tv z 0 LwU.A- 4. ,c 2 "

4' 4 ~ 4-.-:4.44 237 4

0 u

.aa

w at U. C.

u ue 4.

LC C 4.v*6P

A 21 c cc*. . .1 . 10 0 n

INM A 11 xoa,1 2 6a"1

me .. . , 1" 0" . a

0~ ~ ~ .L Lw L c-

L Ca' C 0 a 0 L V 0 L

- uu wo ,* * 0 edl44IWd d- .1.4. -M' o. 4# e tr .vF

39 44 i - i j Lj C' 0 I .a xx 0 .p V, CU a C aaa C ca0

UL) -' " '

kiI )a lI of te 0 4**-- u a* P.

.4 gi 9A2o0CC ~ U U'L

**ut~ CuE *..*wu .EU *238

c

*

14, 0

* U

14 1 4 V4

-ý 0 9.0 , t, u

-n L. v 0 C 61 .4.4g

4 a' .L U L 4* .iccC 9.z 0. u 2 0 MCI

U5-0L v. a " .&

*0.. L94CCC CL 9'*9 CL

*4 0 L'U A 4

c-, w94. a. ' 9 0945L 0m L , L .00

C* C' a4 00 a' .. 0 009C c C U 40 Z.9.. .4

4 L C 4,' 4, 96, 2 W4' 1w0 9, UU 0 v 41

d r. L£,. 9399.9f 4.4r00V.4'9'4. 00 C 4 'a

.9..04 19 0' r * . Z .~9CC r). 00 z4 0 .9. 44 x

o 1'-y9..9' *4990L0U 0 04.,.,~~~~~~~ A-'). 4' 9 -3. ~

C '4'4'L 994 0 0.0 4*9.9 L 1 VUC .u z a9. In

a~-"" V4 L9 09U ' 03 59c9 LC. I*C9, 34990. w.*..99 C 40 - -~~~*~~~~t'~~ 06c0. 0' 0 o L S 4 4 E

.. c 4 c9 12 .1 . 3 5OC 3C c fa c x .1 00 99

- i9.. C '.4 C' m9 t 2U5 lI 299 a 000L .U* . * . .4 '0' 9'C C . I 0 IL 00.4 .4 44 0 6- L C t

c...~,. q'- I- CO 1 B Z 4. C,9 11 e04 C' 0- 9 4

C,- 644099 00 9 -&.4 a 0.0xc - ca CL9.9. L a. f, b.' 0 10...9* @9.04 0 'Q6) z . 'lC 0 *l'999 L

C 9 m 4IL. C1 00910

c2 a6 20.. u*0 .p 4rV'9.' 0-'94 L 0 9 ' ' ' 9 C'''0U 4-00 94- 0,V - If C 99 ''.'

C C C C C C CLC vCC C C C C C 2 2 2 ' -. C , 9 U 9 9 0 9 1

~I 20901. -1 "' -J.J U U.

239

.0 0

4 c

L

c *

4$ 04$L V

.4 J3 C,

4~~~ c~ * e '

ID4 034- V c L N4 c V*000 OCCO .. C

a .v -LLv , EuC

L$ V ., a U. c.C 0*'4

*8 4 e T o-IV Lp w. pop Pp .

a*4qCC C3 10 2.0 0 0

e. -. 7r 0 c *0A0C 4- C-i 0 .. a'

0 505

I 4$I l 1 4, 110 0'0' "' e0 0 10 0000 l C 0c ~ ~ ~ ~ ~ ~ * L c c uccc .c-. cccc -4.11 . rrcC .

*0 0 0 0 D 0 4 m p0&. 4a

14 ~ ~ ~ ~ ~ ~ ~ 44* V. 14# 4ý 4-ý 1 ,ClL C' VI 1 C 0ccL0~~~~. 0- 0- 0 4 0 4Q.UuVk 4 cUL ju0

240Ct - . c

SV

0 ý4

- - .4*L a

a 0

03 V 4M. MC 47 0 M M

us. W. V* 0L CL

944 L 0 x- C

L~~ 'A11c

2a C 0

o4 a A . 0.~ a 14 Cý&a 0 "1

C.4. IF . r

cc a U. C CLr L LC.L L LO 5C

c L~0 L4

V- v..J -'. V V a V -

22 Z4 =1 .c C c C C C" C j L~ V CC LV

c A. * "F * 0 ~ a4vC, ~ ~ I'l 00o 5uc )V

L1 4. fC 4

~ zi. * ac ~ a L *2*o1

4-C

u* 0 uO u0 0 w v

W, u4 4ft UU0 I It; I00.C c c c c

C)

K 0 0ota M U

aC F.I a4 0 0 a aaL- 9 W. %.% l%

*, wc 0e .a V -0

I.- I-U.1 0 - 1- 41 u -# 4

%U WU cye e0& 0w U&

* Or2 Ce 9A C 0 J J 4, A eL

c 94 00EJEmiWMe..4,.UL " 0 1 a CV' L U '4 C,0

44 0- tee4. 40 4 0 -L 914 v .42I4 4.4 - 4 V' a 4 4+0 I I

u '~ CL vE u 'a41 P Ct +,00+

u40 cc 41zi:I 4r

e z -.eoL4 ft a Ca V ,C,

'0 0V 402 41q4 LVl0 'a. 0r 4 c 0'0 L.C X-

INU. I.00 0 C U eo c a. 41L~~ L 0 S'0.. 0UOCUOL~~4 'a.4 IO O VL U. L 2 4

C-U ~ D O C4 eI -.u l j e u 0- .0 0~0.4Urq.4u 00L @.-4I - cVI

s 0zI jI.1 0 u a~o 1 *aaM -- 0 -4.L E'

AM A A. lee -vie.C ('-.-iI 92 . I f 0 00 .am*0t.260f.I

LC 3r &ws L o eca tu zA 1 w LA. -9 e- j . .i u -c 0 .0 El ' c 0 L 6l c vW 0 m

~.4 V -2*-! " .- C~ 4 .. 4242U

8 L .4OR V A, C

v4t, C Cr I C

a C

0 .4 C 4

0 W1 a .4

a 00.4a

c an Bo o -U

*0 L- In 0000

4:. v. D . LO .04

C t.- A0

*0 a u .g1 0 .4 a-4. U, v.

Ut g A u 4- 141 0 7

4 g xC4 *

co oc e'* - , £J%

110 *.ar 34

* 'D0 * 0 0L C a* 0,I0 L 4b. A aCO

4:.0 CL 1 0 'A W U V 20 0.4 4* 34'3C L RIO 0,@ U . 9C 0

I A a I -44 CC1 1 41 V 041 3.

* --- 1.4 4 1. in.INV+ 4. 3 ý U00 04.' 0 .1 vf SO IF4 * C U9Ick0 0r

I ~ ~ ~ ~ ~ * 1V VVCv C .- .C 0 0 U .. 4 I- .r 0 M. r0 V~ i if0.i t 4. 0 w 004ý-1 .

v: 04 o1 20 4

*C.4t- C 03 04C) . Im Cceq -4v . 321 9. V0 +01

&3 c.* '04 4..

*m ~ @L* C0 0 a.2 1 0 L vD

*04144V *0* 0 . . - * %. A I ...a C'. M0 0'CC0'4 M S %s A 3.4A0L L

C' VE A' 4 E.R 4 f I N W -4 - '-t no 0 0 A V a

a 4. a 4 41' 0. 0.4 It e. 04 V. 4 0 a4 0 4U0131) *-1 046coIJ1jWQLýu'S 8040 Cc" .S ~ *j 01 W~ 10 41-. w 0j

490'. - 4~ .o w 4. 4. 2430. 0 4 e 0-O 440 4

0 1CL

CL

00V

.4

4- c

.00

2 12

OwI

-- .z 6 "1 -

-. C C '0% In a -NL

Of U' UV 'aVV -

S1 OL 0 . . -- ---C L 404 4m a 0c&C . C.0c c £ 1 0m"c01

.4 4. aL n :aaa ý m a w , a m a A M 9 0 a ý - 0 '

£4244

w C, c. v n0 q a -v v a C. V a c

C 0, 01 ; 1 4 a :1 . M.t * 'o "lm e O* 4 7C 2 11 C a. C 4 C u 0 C 0

L L C L #I L *I C L 41 L. Vo~-. 'o. . .- cC . 4 0 &s .S A 0 & X 0 4 a ~ ' a eCV *~ ~ 00 A 0

A. ' .v L a ~ ~ ' O. vz O' 3 04e a4 A2 s 44 .A . .a 0 0, J31 0.0 4,43 wt A4.31

* 4- .o f a 4. 1 0 A 0 A 40 a a.. *U ~ * 4 P4 A ,4

.40 0 2c v 0Q 0 PCA 0 Ut0 0 'n C Z U t S PC 0 Ut0'woow 0 GNU 0 eg a~ a . t

245

IL 0

m IL

CV .0

L 0

M 1."C.- 0 :

W. 9

0 n 4 r--r0 0

W,~ 0.x-

~~*.. Z .' P

1. v c L+, Cc L C L C C C C

*.-.,a~~ C,0---c 0 m ~~0 aa0 1INm

.0~~ ~ ~ ~ ye. cm- C-0 ct ,~

49.

9~. P0 0 0 000246

4, Q Q w 0 OL

"0: W, 3, a P. 4C 0: V 43 '4- '0 C4ý

- C to. '04 1*, 3, 'C' wb Lo '.2C: a4 c fr X% 00 - c . 4 -6

L L', C 14 L L- C~ ... .. ~ .o E . L ~ ~ ~ . 6 4 . 63'~ ~~~~~ LP6 3"-eNC...* . . 0 0* C C 0. 4 1 63 X'~ 0 0*~~~ .. I C'. A.3. 041 A~f VA. CP. *.. 66* 013' ... .0O *"X,, r .. ... I. *'I

2 f ~ 0 * * 0 *gO N 020 0 L.0 0 02 00 It2011 cV J a0' :.~. a.' 3' 0f.2U A'~ A93 * a a ow *A c'J 0.4 0 O'1 .0 MC 1 .0 c

247

C C

C,.

4'

.4'

c rC

*1 14

I* .. T) .. v .. V -1

C L C LV C c cL c-

4~~~~~~~. 0 t- 0a.

zr 0, v - Vo 6 A 0 0 1 +1 0 Jle 4 6

.1~~ ~~~ ~ ~~ ~ ~ ~ ~ ~ -' 11w 0 m4 )43.0 0 7 4 N ,sa0 o

tj~~~~~~~~~~~~ ~ ~ ~ ~ ~ ~ c -lA C:Olc0Arc uc0a c uc0aac czvc o 3V

248

pL

4*

c c c c 5*

11 0 V. o Ir cc-j j ul 1 r , Wr C 1 41.1 cc 0 cc c c

-r Cýc OC a *

WI LA *

0 w 40'- P- le fa r

v O,.-a 04- .4 a, f40,403w 2uu.u u14 w 7 : + UUUU

L, L -L , 6

0 C3 a 0 *cg CCCC4.. 444 .0 .04 -1.

4.. ~~~ . .. .: ..au .U . . . .

If 00. 0 A4 * ix LA *j 0 -U a4 4. o

c v 044 c S- P. 0 441 04E.- *4 1.iu N 4

*~~~~P 'A 0 0 *2 4 .00IV ~ ~ c. CCC -f 10 . 0 1 "

c ~ C.-~i'44 I .NJ * t.b . u 'a z

4 00 00 0 4F M 40 4- 8 0 0

C~~~ 44.4. U~O4.

*3 14L A 14.- 4C 0

U. c . c 6- OZ.4 *0 £ 0 ~0.J1.La .L +1 a' 4wc 4. r C. auc L' 't Wta,0 oz vW&V

C, .. jr.. 3 *. a. a c 434 0 tojma it,000I4 r. -. 4.4O4 -.4 0 . *444 44 6

I . CC 1 0t~ -4 to 2 -A 44 -M* 4, 01. Ib I fu. I.- 4 V 8. C 44 w.A3. *L-"1 44 4

-9 .3 qw P- vi 14.4 0 0U 22 -4 .0 -.CS~ v IT 1.'

VI .j: 4, 0 M U 4- W 0C a u.L~F Id A. *. 0. $- *7 0-W

. 39 0 414 C A .I

u -. -j.. . . .. 2 m0 N . .1 31a3rt* .1 Cr ELM p. c 'a 44 V.. L P rL.

* 0 &1 01 lb 4C-q

*~C c1

L 4 1 c. C- C4 C V04.- 0 4 0 0 4

L A. C~r~ ( * 4** 2-1. 14 .4 4' ft Oft 4- ft'4 O.64 0

IS I 04 0 - - 4- 4 4 I - 4z.. * . - * . 0 I ~ - -

4i.L 0.S 4 L "0 c 44 4 q- mf 1. 4. '.3 . a fli i4 3005U4 4IC4 a- 00 - 0

*CN '.. 4' , 4, c I a 21 on .0 lp B.0.) '4 ?1.41 4- x -4 0.4" 1141 -4 :.0 34 : .o a0 ca 04 0 cc 3 0U

'0c604' .0 016 a. -9 . 0f I.- E.1J 0 M..' 414 U;S W C '> o , f t 2 4 L2 1 A vv v44 nOC.C tj.444 Ci u4L L *4-U CI.UC .44- f c

08 LO LL

222 -J .~ 249 S2 2

*

4,

* - C

C:.

c

C *3

C, C. W

,A rr

~~,C. 2 Ca. C.

f4 In

Irr

-b~~~~ -A0 a

IB is a a~.

c 'cU soC.:UC 0 AS C4' 9~, ki- C' no' i 1 4ID F ,O,)' b,"o1i U 1:jju0L

250

II

I r IC Cc C uLC

C. 3C' 0 C 0 0

orp~ot.'C tr r 11

vu" L, C.%Ja

In C, 4r C.C

e - " c r * -

.44 10CO

-~~ L~..4tJ It OV

o4 -- n 9'.,-.a.-4£3~~ £3 0 P- U£ 05C

ID V5 04 4 000 ZX 2.£C'00* 20 (2 - 4q .p LCI

C. L 'k C

CI L '0 $4!) ' nv 0 V, V

0.~! L0 v c*4 C'C A-:1 rrI4 CC @ C 16 4 U U'

*.) IC II 0- 45 c4 0- 111 w.ca4 e UC3U V C QS. M- 14' C$ of&). o L3 9 j-0 # nac;L .

.~~~~ ~ L4 L54 5,. .

C ~ PS~ 5 4.4".~ 4O LO251

00

4,ý c

z r -4

V. LS.0.

C,~~4 L ,07

V 3 a . 4'

cu-4 vtc P: Lc rI ,

c P: CN,

m4v'Uq. L W C CP C L C '-C C C C C 0 CC C C

Vc I.m

2525

it4 44

*I, cz

c . .4 .

it 0 -1. 30..4 . 1-

+*0 L W 2 C

-,. 1.10~4.,- aC.C££C C

it CL a3 a ,41 r

* * 0.0.

L4 L L L

It 'o

x0 0 'a. a

20 : a ait c0 a 94U

*4~V `40 0 It *4

CL .4 X A' 8

'0 * " :,4

*O CC2 -20CL it q 0 ý 0 4n1C' LL L V, .0 C, .0.4ý ' D V2L i . m f

C! *00-~U -' 0*0 0 0 . 4 .0 a a. .. ..a~A -4.00 0 -'.4 L4

'tVavI VD' 44 0 It, v In S v ~ 9 00 4 v03x' 0 1'o * v U ox .4 'D .

U' r4 c 1 5o a - aca c C A 0 A1 8At

44 a a & 0 o a .aa0 ac " : I so0 S. .0 z vo- o- O'c0

It 3.- a 0 a 3k a a *sk CUjui ~ v uQ* k..--. a .4 .0 u C 04 a Lc0^aI I-*0 0 £ I- WI- t I- I

*~ ~ t- L L U44,.. L' C (

0 .4 *.4L 4-4 ~ .4 £253.

m0. 10 go C

C I

a 06

m 41

C vC

rU I n C c C 0

a a c W. A I)

u C C V4. If*800 .44g -ý am 0

w u u Q, c c 4a 0 ~ * a' 04 9

.r -0 -V 0' P.. tj

*. ftC4 ý -- - -L.OsL ' l

aC &1, ,.L-g LO -l -a Cc -

-6 '8 'o m Iq

1 C C 4?.- 4 sl -1 - - -1

841.4 244 -u*'01 88 40 3.-

6 fI V V ý 0 1 (.L*~ 0 L Ap

CC L.CU*CC CV' C:r*UU'- T'-I C.--

-. -. I- It v = c o T I-0444.. 8 %0,8 N4

,

ch ~ ~ C.CO

a -" 1 0 'a v V 'a C L 1 C C L CC. c rcL 'C' c I Ccm 25 c'I - - -*- c

If on a- I.Ia'1a 8 -164

9'. 4. mr)0N0¶8 u.4. .J m U. 1aa- 0WM2,- 4%C *.j. C' 18 c ..V D c9t*868 10, S3 -O 6 J%

C 0..4..-4.254

C- -l I- C,- U100f

a' 0a a 0

I. v. m

L 4, 0 '

010 c Ca a u0-0

L. 4. 1 0 0 LuC C~ a S

- -. Q "I" '

0 ! .4 ý 9'

.* .. . ** t,S*4 -. 09000 1

4' 0 V n IL~~~~ CC-V LLw 2z vz ."00

a- p : mý4 a 0 a= 3 w3rjj -. + nC, *A 0,jo. .. . . . . .31, c 31 0 s N 0 20 an3k e4z 'o C, - +' .' 9' SA w U,

*A a -9 P ,Aac5wo cC cc

2555

Macroscopic Material Properties Task

The purpose of this task is to determine the mechanical properties ofLCPs to be used for the design and analysis of rocket components. Theproperties measured are tensile strength and modulus, compressivestrength and modulus, and shear strength. This task is using thematerial's mechanical response to applied load to study severalcharacteristics of LCPs. Effects of processing variables, materialanisotropy, the "skin and core" effect, fundamental material behavior, andpotentially annealing behavior can all be quantified through the material'smechanical properties. Work on this task up to this point has focused onmeasuring the tensile moduli and strengths of these materials in anattempt to develop usable processing methods, examine the materialanisotropy, quantify the skin and core effect, and develop a materialproperties database. Future work will continue to examine anisotropy, theskin and core effect, and the material properties database. The effects ofphysico-chemical annealing on mechanical properties will also be studied.

POC: J. ShelleyOLAC PL/RKCCAEdwards AFB, CA 93523-5000Phone: (805) 275-5394FAX: (805) 275-5144

256

Ul)U- M

am 0

S 0 *I.J.J

C.0.. ()

o) Z(w*)Ea

E 0(E EM)LL

o W aa)

Cu C

257

4)U)

U)

a)

a) CL0 0.

0U) 02

U)-

CU)4 U) cuU'-0cc) Cu

a) cu

U-4 J19) I

U) C1 t:- a)

cc 0CO ) 0. C

0 )LU a) La.0 Cu- 0z cc0

C 0

CO)258

CZ

cCý

cn tL LO c ~jL

CL/

CoD

*0 CT$ *

'10*

o~~ a CfC0 N L6

U-z

000aC) o

ca Eý cu 0 ~ , - L

00ICL 259

U)IM- CuU) .04)u .2.-

0. 0 U)

on _ .

Lm0 0

cu 0->%ii

C))

a) EM-0

o 0 Cu

U) --. -.40 4) ~ a) U)

oc 02 w)0.am

260

I))4)

0

o E0Lo A C,)

EMEo %Immo

0. . C2• - (.) a

""--0 -.USEEM0, o_ >% oo - " SE M 0 Wilm 0E ma 4D() Ic. Mr

C o o ED c 0)" L 0o = V) ) 0.- cLo U) MEI E

m% ccc

C,))

MEEE- 261) Z U) j

0E E E0) 0

4) 0 0D

4) .cc) 0.0. E0

U- 000

0 0 - owlCM

a) U) 0 0 M.

Cl) Cl) 0

a)II...262

0.

0

99' L

LZ' Cl

tz L -E

U) 0Lo . .

% %z

0J LJ . . . .

0 0

0<0

L*O

8- t

(!sdVW) L L3

263

0

o~scaCUse 98C

EL~ In %

%6 % % %%%

C13

GE L9 En

06Z6ZL4

01 W .. iC., N E~ N

-L C

0

(isd qi14~i CVW!If

264

Lm 0

I-t

LM T

4)

(0

0) o-Q

4))

00 w05 4

0) oo>- 265

0)

00

0 >0

-5 E

E E 0'EE

*0U)*

U 0) a)0?0

266.

13 C02 C4~m

2U I

w*o 0.,k c Icc 0=( G

Cul

00

2CU

0 0)

cc CMJ

CJ) 26'

CV

o2 C4EChm

OONNW%

0A 04 oowu

owo

oo at o4I zc

02 C0

CuE

0 050u)

(a)268

* Thermal Analysisin

LCP Characterization

Thermal analysis was used to study the effects of annealing on a varietyof injection molded HX4000 samples. Results by DSC showed no strong ordistinguishing phase transitions occurring in the heat flow curves for any of theHX4000 samples run. Results by the TMA showed that by heating the"annealable" HX4000 for an extended time at a temperature just below its initialmelt, an increase in melt temperature then resulted. These same HX4000samples were then run on the DMA. Tan Delta temperatures and peak shapeswere found to be changing to the same extent as the melt temperatures werechanging as seen by TMA. As noticed in the DMA results, as the "degree ofannealing" increases, the first Tan Delta peak decreases in temperature andbecomes more broad while the second Tan Delta peak, increases intemperature and becomes sharper. This may be due to some sort of a reactionor break down occurring within the LCP backbone. Another thought is that thetwo Tan Delta peaks could both be glass transition (Tg) peaks and if true wouldtend to prove that "ordering" glassy-.smectic-,nematic phase changes could beoccurring during the thermal treatment of the LOP. Future plans are to run theTMA over the same region as the DMA and verifying the Tan Delta peaks asglass transition or not.

269

000ru

0

Z L 0 m

0 .... 4tC.-CLm (a

Cl 0OL 4)

0-JI

aD 0 l)u

LL 0 m

-0 .4CU 0 ):3

0.0zL

CDI3O

ui CuHZ

Wz

OL HC

0

HO H

> - -1

ol 10 c*N4J..5 0 0 0)

270 0- I

0

Cu

* o c)

0 z

0 0Y

')O <0 L q-T0

1- 0

0 C0

.I 0 cI 0 0u

0) L 00 a0C

- 0 cc

00

L

Icu 0

CD 0 0-

:3 cu(L

zwiCD

U) 0wz 0

I--

o wo

0)

0 Cu0 0 x Co*0

IV) L') CL -K

X co 271

cu

tui

U)

0 00

M ** U"

a)L DC)C''0) ac 1 0 cum4 4.*r w CLr)j

0

0 L

0)Ea)H-

0 3:

z0

z oz C.3

Iq0 ZO

*qx 0 r

xom

r4* .. o0W 0 0 0 000a c icE 0 0 000

4N-a-CS 0 0 00

Wr a6UeLlo UO~suaw~c

cu

41'

CL

uincu Z'

00

U)'

0'41

1* 0 ECD bQ

LL 0 :L

< CD0 L

'44

0 M.

LLI-

0 0.IV-t

10 z .)

C..a) 0

0 x.. Cii

0 Z1EC 00

10) 0nxC.I1f)8UB404*SBWLXq4273

0u

00

m C3m 4-t

LO.631xcm

0003

zm O0 4J

V to

CCD

H -c

0

LUL

CO CD

(0

w:3 0

0 * CD0 L)ZC

0000

N4 S

c' aa I -r T- - FT-F ' I

o 0 0 0 0 00 0 0 r0 0 0 0

2 aJ ' 0u CuCD030

X L)(Wlil) abuoijo uol9Uaw'cQ

274

000cu

4-)

0m C

cu4

LLIn

001 01x

z :3

000

LI0

< 0 4J4 4 10

LL 0 0 r

00 :

00

C.)L

0:

0 '-4

.4 I

w -0

z c-.z

0 ZC.U0000Iq 0 0

jY) -4 -

0 04 a 0SE 4I fl(wri) 86ei~ u TGBW

275

0000

Im

0 4J

r-4 0 1c

0~430Zm 0

1-4 0

co CD Cim m~

CL

00

0

10UlX 0S

00

030

x ul(wY ~uL.u~s~:c

o 276

0

cu

4IJ

0W I

to4V44

V44000) 0>

Cm ')

-'<Ia

:('L0 4

4J

< CD 0c c " -S

L 00

< 0o

0 L.

010

C.) WCc m

0 CDCu CLz cu

0E~J0 Zu.

cooDC

o0)0 0 00 0 0 00cOor) S0 0 0 0 0 0 0E N4JE Scu cu CD c 0

(;;;,,) a6uel43 UOTSUGu.;TO277

00

cu

T orccu

z c F

o:to

I-4.C L404)

~L a c

'• II.

"0 "

{1..

1j 0

go C3d) L CD

IL0 MJ 3r

00C

L3I: I-I u-I- .I- -

a-

12278

000c0.LD<L

CJu,:rv)~J*H

NL

z T Tu

0x 0D.2 0 00o2wN)C UCxf~l~

C27

0

IIcu a

0n0

-0 1 0>

TZ a) (1,4c

<*0 0

03 L

CL :*fU- 0 0

:9C.0<x

4

CDL0~ ~4J'-4 go

L03

0(0 0

CD

a.

o L)

U3 z <x

z a) .1-

a~E00 0 0 00 m] 0

(Wn) 3~etZ)uo.~Buaw.jo279

00

cu

0 D

N-w cu>m3.01

-i U-

2 4c

C3 0

b4J 0 )r40C)

0 0

'-4 0.c

< 2

44

cu

0

X 0

w 0.

0 0oN1-4 0

0. W0. 0.ce0

No z1EC u oc

-4 ** 11

u)cnx . -Tri T6 e~ U TS 1;(

280

C.3

0u 0

00CU0 000

Cu 0

4 0.

0 4J a: 4'

w 0 (0

I- 0 c

z 0

0.0LO O

SLo l 2

x Z1

wL wa) CD

CU0 S

x .40 Z

cu 0

Zx0

000.0

0 0 ) W

0. r 40000 0I C0O d0

Cfl fl <x;~ IVJ x.

x m m2a:

W cu

cu(UdW) .9 (---

* W~I * IM Z

0 :34I

'4J

.w 0 0

0 4

0 .4 0

)Lz

4- 4

cc ICD !

C3,0 CD

o2 Z to -000

:3U 0

In z

*0

O*CL 0Cu 0 )

0 -000..w

00 C.) C'00-Iqx.

T 1 ~cu

CL a r-E '0 0 0

0 -,4~ 0) 0 00tn cn m:ci -,aC j.

282

0u in 0Cu -4- r1 U, 0<

-=

EO 0i 0eW 0

4W Cu 'L4 0

L "0EL M L) 0

a: CL

C..

C3

E 0 0

o 44

T0

IL CD

0C 0

10 z i

om '-u 0C) 0 cc

0o z000.

m win -

00

00 X'. 00)C ~

Cr400000 0 0 0r0 U)CliU

qT m283

0

(OdW) 9i U 4

to >U

CD C3m 0

co(GdW) .9 ---

-11 .

UJ *0< M L :3 0

a 0 C0

oL 3

ILU 0

33

to z

00:

00

rI0) C)

4 284

0to 0 in oCU cu 10 0<

4. 0

V.4

Iý *n 0

4010 mE: O00~ <

I q

4J 0 0

ZM L :3 C)

LUL-1 06

CL

CL CL

0 00 Z

m 0 CDn - 0

x H-

0) Z

* x

CU 0 U'400 -1

IT () wi

.. 0... - 0

o *0 ca o -0;LU C a.~l3 a ) 0 0 00

285

X-RAY/NEUTRON REDUCTIONS. D. Osborn

UDRI, OLAC PL/RKCP

The task objective is to investigate advanced propulsion materi-als for solid, liquid and next-generation space flight systems.The scope is the support of the Advanced Polymer Components andNEMESIS initiatives. The technical requirements of this taskinvolve the development of algorithms to reduce synchrotronspectroscopy data.

This paper will report on the investigation of the state of theexisting analysis tools and techniques. It will include a dis-cussion of the hardware and software used by members of theresearch team. Platforms will be discussed incorporating theseresources to improve the data analysis. This paper also presentsan overview of our international collaboration, specificallyregarding synchrotron and neutron research. Finally, stepsinvolved in EXAFS data reduction are reviewed in order to give anappreciation of this highly complex and iterative process.

286

0 0

w~ 0)

.00HZU

Clo.HC

wu

287

C/)

00

00

00

L---J

00a--

CO)z

288

C/) 0

z V)0 Cz

< 0.

DL 0Lw0 0zC

CO < Z aw WAJzm ocEccI 0 mr

LL.H(! Z.L _Jj

-w w w D 2:3 zC

H IJo (! L. - Z>-I<& yc

C) -(. <co cc F

LL oFo<DZ-oLL4LW-(.CL0U=~ J LWF-M M xW289

0

I 1 1 1 1 1 I I I I I I I I | I I II | 1 II I I II I I I I1 II I I I I I | 1 I I II I I U )•

"ClJCL,

CL\*--4

-r

V. 0•Ul

_I-

ID IX~~~ 1J1 0 inC ''Di. ID S, 0f f'' f'C 7'''I) ID Sf' ' ., C -) f"-: -n D I'0 If) -D

C.J I i-; -1 GO T, IT, o -0(0t--. t--- r-rC, U- 1 -I -- r 7" 17 ) C'' A(A - - C)

290

CIu CL

1T3

VJ~i

ui

CL 0

al) 1-- 11 r -a

r 291

Il-)

-0

C-1

U-I

o Cl0LU-

Ln

w f

CL

wz

0 DC Dr-rC DC)r- m ,In" D1- lC r--

U- I'I' U''1 I U-) ,11 1 U 11

- ýr7rT T, 1:11 CID t--- t-- I-L I£1 U- U-1 -; --r C1 i 1 17: - - -l I

292

C' l0~0

I-I-

U-14'-

Li-) N

293

-UJ

(L

\ N

bILl

C-D C ED rl C)

294V

I-

'.7.

-r-rj

2r9

-j:

, I I , I , t I I-1'- I - I - I - ,I

0295

W (70

LLi

IT

f-o--

0 7 . . . ....... _.

-I

+ + I I I

S++.1 :': H]_-6

296

SW(-I-

r-17- CI

X IHJ

297

II * '"

• 1- + /

II... .-. . .

"i (O

C..'

.,-: : -- ___-

II--z_

If) U i L b __ L b

298

CD

w

* 012

3oni ij~jv3A I V 1-,

299i

-J-

CL

I~ HI_

00

TASK #15: MOLD DESIGN/PART ANALYSIS

Thomas A. Elkins

OL-AC PL/RKADEdwards AFB, CA 93523-5000

(805) 275-5303, DSN 525-5303, FAX (835) 275-5144E-mail: ELKINSTOPL-EDWARI)S.AF.MI L

ABSTRACT

In these times of diminishing budgets and manpower, understanding the behavior of liquid crystal poly-mers and the process of injection-molding is critical in minimizing the time and money spent in developingmold tools. The Phillips Laboratory has acquired software which works with our existing analysis tools tosimulate the injection-molding process and optimize cooling line and runner geometry to minimise the cycletime and part warping, as well as indicate potential problems in the part before any actual machining isdone.

INTRODUCTION

In 1987 the Phillips Laboratory (then the Rocket Propulsion Laboratory) purchased a software packagecalled I-DEAS (Integrated - Design Engineering Analysis Software) from Structural Dynamics ResearchCorporation (SDRC). The package contains a series of modules which share a common product database, sofinite element models may be developed from geometry created in the solid modeler, engineering drawingsmay be created in the drafting module from the solid model, test data may be input and stored with finiteelement analysis results, and so on. SDRC has added modules (families) to their package to enhance thecapabilities of .,eir product and enforce the move to "concurrent engineering." Also, the I-DEAS control,or master, module has an open interface so that users may add interfaces to external codes, a programminglanguage to automate tasks or add capabilities, and a function library that allows users to write programswhich can interface with the product database.

Currently, the Phillips Laboratory has the following familiesGEOMOD 3D solid modelerGEODRAW DraftingSUPERTAB Finite element pre- and post-processorMODEL SOLUTION Linear statics solverOPTIMIZATION Design optimizationGNC Pre- and post-processor for NC machinesPLASTICS Mold filling, cooling, and optimizationTDAS Test data analysisSYSTAN System dynamics solver

PLASTIC FLOW SIMULATION

The plastics analysis module of I-DEAS uses a finite element model which is created in the SUPERTABmodule. The model of the part is created using thin-shell linear elements (either quadrilaterals or triangles);the ntIld exterior is modeled using plane strain elements; the parting lines and inserts are modeled withplate elements; the runner system is modeled with either cold- or hot-runner elements; and the cooling linesare modeled with cooling-line elements and/or baffle and fountain elements.

Once the model has been created, materials mrest b-i- r~vg'. f,, m•4,- .,,,,l, inserts. parl. and conlnnt.This is where the database interface from task 9 comes i,, ph . th, ph%,-,- parmmi,.-,.rt . ri. ,' .vtr'(lwhich include clamp direction, injection pressure, cw,,,,lan i t,'i,! . -. .... I l linl' prcssirv,. .,ling hlie'connector geometry, mold temperature, ejection tcmpcratlire,. ,et,. '11, 'It, i.i -lts (he, analvsis method andnumber of iterations for the fill and cool calculations, and whet hr a; uI;,t, .r,. iý ,, bet created for a warp andshrink analysis. Packing and holding data are then input as pressure arid velocitv profiles.

The software performs a cooling analysis first to estimate the nmaxirnum time it will take to fill the cavityand cool the part to the ejection temperature. It then performs a fill analysis using the estimated time andcalculates the temperature of each element for input into the cooling solver, and so on. Even though themodel is created from thin-shell elements, the software internally divides each element into 10, 16, or 24p (user-selectable) layers and performs the analysis on each layer.

301

When the analysis is completed, the user may read the results into the product database and use thepnot-processor to view various results, including an animation of the filling process. The software stores theresults at various time steps from which the user may select. The following is a list of available results

Mold FillingTIME-DEPENDENT TIM E-IN DEPENDENTPressure Ejection timeLayered strain rate No-flow (freeze) timeFlux Fill time (flow fronts)Layered temperature Sink mark magnitudeBulk temperatureBulk velocity vectorSolid layer thickness

Mold CoolingPART WALL 5URFA-C COOLING LINES MOLD .EXTERIORTemperature Temperature Temperature Rey nold's number TemperatureFlux Convergence Flux Flux FluxEjection Time Pressure

TemperalureFilm coefficient

If the warp calculation was selected, the user simply changes modules to MODEL SOLUTION andexecutes a linear static analysis using the loadsets created by the plastics module. The user can then displaydeformations, stresses, and strains.

NOZZLE DESIGN

The first mold design under this task is a rocket nozzle which will attach to the Air Force Academymotor (also being injection-molded). Figure 1 is a drawing of the nozzle designed by Dr. John Riisek andMr. Hieu Nguyen. The nozzle is 4 inches in length and 2 • inches in diameter at the widest point, which isa circumferential rib. The leftmost part of the nozzle (to the left of the circumferential rib) will slip into themotor case and be pinned to the case. The motor case will rest on the lip of the circumferential rib.

The recommended lower limit to the size of an injection-molding part is 4 inches, so that was used asthe thickness of the longitudinal and circumferential ribs. Also, to reduce the possibility of sink marks, ribsshould be between 60 and 70% of the nominal wall thickness which would make the nominal wall about 3inches thick.

lj

St . . . _IL. . . . . . ..-

4 ,,,/ . . ... . . .. -, . .. .. .'3f

302

MOLD DESIGN

The nvrrnll tlimensinns or the nozzle are roughly equivalent to the dimensions of lite 2xi molor came.Mr. Chril F'runk from McCl,'llan A FB sntggest'd timing the same mold base for tihe nozzle, thus elintinutingthe cost of having a new mold base made tip; the core and cavity pieces could be machined in-house. If this

works out the same mold base could be used to make either the 2x4 motor case or the Academy nozzle.

PROCESS SIMULATION

A finite element model of the nozzle was generated in ]-DEAS and a copy was made so that a structuralanalysis could also be performed. Cooling lines were added based on sketches from Mr. Rich Griffin at Hill

AFB and a runner and gate system were modeled. The drawings from Hill did not provide enough informationto accurately model the mold or runners, nor was there sufficient data about the injection molding machine

at Hill AFB to properly set up the process parameters; however, this was just a first attempt to see how the-

software worked. Figure 2 shows the finite element model.

-;..] , - "C . ' p -

-' ... ._ ; q .-.- ,

• . - • • It " .• ..

C L r: L 6- -"-1

* I-, ' • r ,'T| r ,] L ,;I- .:

Material selection turned oul to be a problem, because there were no LCPs in the supplied database;however, this was only a test of the software, so the materials were not that critical. The materials selectedwere P1`1T for lhe part, 414 Stainless for the mold, and water for the coolant. The default process parameterswere selected for this test.

After submitting the job, the user may request updates as to how the analysis is progressing. The

following is an excerpt from the status display.

Step status at 00:46:00 --- Mold Cooling --- Started n-rrv-n n".:rn:19

Current iteration = 3 Total CPU time 00:38:16Plastic calculation complete 00:00:36Fourier forward transform finished 00:00:48Circuit calculation complete 00:00:10

Mold: Matrix and known vector formed 00:00:13Mold: Linear system of equations solved 00:00:50

Fourier inverse transform finished 00:00:26Writing to the print file finished 00:00:05

S303"

Restart file write and other tasks finished 00:00:01MOLD COOLING STEP COMPLETE--------------------------

Partial results for iteration 3

Ejection time = 11.6146 secondsMaximum deviation = 1.0339 C Root-mean-square error = 0.8347 C

(3,10) 2-Fill. ERROR: (CVDRVR) Miscellaneous at timestep 70Short shot; remaining steps in coupled analysis not executed

Step status at 01:03:44 ---- Mold Filling ---- Started 18-FEB-92 00:46:08

Performing a IOVISOTHERMAL COMPRESSIBLE analysisTimestep = 70 Process time = 11.619 Total CPU time = 00:16:50

Percent filled = 71.6593 per centFill time (requested) = 16.0000 secondsNumber of nodes filled = 362 out of 576 nodesEjection time (computed) = 11.6195 secondsMOLD FILLING STEP COMPLETE-------------------

Pressures (RUNNEI1): 16.3 MPa (inlet), 3.4 MPa (gate (min))Flow rate (RUNIER1): 4.09E+03 .uC*e3/sec (inlet)Temperatures: 268C (melt). 264C (max), 221C (Bulk (min))

FILL/COOL RESULTS

The next two pages show some of the results which may be displayed on the screen. The first one isfrom the cooling analysis showing the estimated time it would take for each element to reach the ejectiontemperature (in this case 150"F). The picture shows that the ribs reach the ejection temperature ratherquickly, but the throat/gate region approaches 12 seconds. The second picture is from the fill analysis andshows what elements of the model were filled in the time calculated bv the cooling analysis. The softwarewas smart enough to realize that the flow froze before completely filling the mold.

03O4

|r

--.p

LD C

305

a 0

C7-

C-.

.a- -r

sCl

L 0

~ 306

STRUCTURAL ANALYSIS

One other possible use of the nozzle is to test the ablation properties of the liquid crystal polymers asmolded and with a CVlJ coating. A structural analysis of the nozzle was performed on the second copy ofthe finite element model. The model was restrained at the four pin locations and a pressure load was applied

to the element faces of the nominal wall. Material properties were input from supplier's data sheets, whichassume isotropic material behavior.

A simple one-dimensional equilibrium flow calculation using the TDK code provided the pressure profilewhich was used. TDK prints data at user-specified area ratios (area at point of interest divided by the areaat the throat) which were calculated by finding the area of the circle defined by the centroid of each "ring" ofelements. For this analysis gaseous Hydrogen (GH 2) and air were selected as propellants, and the chamberpressure was set at 2000 psi. Figures 5 - 8 show the load profiles calculated by TDK. Figure 9 shows thefinite element model with pressure loads and restraints.

LCP Nozzle static firing (GH2/Air) - ODE solutionODE Pressure profile

o

?' I

5:

Area RolIO (neg. - subsonic)

Figure 5. Pressure profile from TDK

LCP Nozzle static firingGH2/Air - ODE solutionODE Temperature profile

C--

It

-P8 6 -¶171 .16 -us -I0 -1.8 .6 185 o t.1 ?. 1. 6 4 ¶.56 a11. toAtoml Ratio (neg. - subsaonIc)

Figure 6. Temperature profile from TDK

307

TDK Results (continued)

LCP Nozzle static firing (GH2/Ad -ODE solutionODE Velocity profile

o- ---.- -- ---o-- .. ••

II

I!

0 I I-r -.. -..- --- - - - - rT" •----,-- -'t-- --- ,-fl.6 -t -ST I -. l5 .51 • -5 .1 - . 6 - .•,1 0 3.6 6 7.6 •1 tst 6 11 t.7. 0

Arc to ', all& - subsonic)

Figure 7. Sonic itioUiLy fhrough "e nozzle

LCP Nozzle static firina LGH2/Air) - ODE solutionODE Mach profile

o--

.-- S

Area~~~~ Rto(o)susonc

I; /

I-

Area RIaIIO (us.j. - ouSion0lC)

Figure 8. Mach profile from TDK

308

STRUCTURAL ANALYSIS RESULTS

The lant two pages show the displacement magnitude and maximum principal stress. According to the

results of the finite element analysis, even assuming isotropic material behavior and believing the manufac-

turer's data, the nozzle appears to fail at the pin bosses due to the nozzle inlet flaring out and buckling thesupport ribs. Tests done by Mr. Chris Frank also show that simply pinning the part will not hold.

CONCLUSIONS

The software does show promise that a part and mold could be designed and tested without actuallybuilding the tools. A better approach to testing the software would be to model the 2x4 motor case moldusing the real numbers from Hill AFB and comparing the analysis results to the real parts. Once thetwo match well enough, then a more thorough analysis of the nozzle mold will be in order. The materialproperties of these LCPs need to be verified so that a more accurate analysis can be performed. The software

has optimization switches which will change the cooling lines and runner systems to minimize the cycle timeand warpage as well as recommend structural changes to minimize the amount of material used withoutsacrificing strength.

C I ULi-. 1 t L _1"l ,I d t L k I L.P I cR e 1 4j '1

rp's .l. IND..10kOtrI

1hi *Uc 9 1117 EIiT( EirhIHT 110141I OP ITPU TuhR4L ArPL',SJS

309

-- t

-= =a--

1310

,c- =-

.r. -:

•.) -.

= I"-- -

=.- I

.I_-

I

* S I..

•oN

APPENDIX 1: ANIMATION

The nnimatinnm shown at the sympnsium were all generated in-house using a comhintion nf comnmereinisoftware and utilities developed in-house. The following is a description of the animations that were shownand some of the steps involved. Except where noted, all animations and drawings were generated by Mr.Tom Elkins.

S inning Phillips Lab ShieldShield model was created using Wavefront's Advanced Visualizer by Mr. Russ Leighton at Phillips

Laboratory. The gold texture on the back of the shield was actually an image of flames that was mapped tothe object as a texture. The shield rotates 360' in 100 frames.

PLQjec Zamfir ShieldImage was generated "by hand" using Deluxe Paint III on an Amiga 2000. The "motto" is a perversion

of the motto of the U.S. Naval Academy (ez scientia tridens - from knowledge, seapower), and does representthe feelings of the author!

The original drawing was scanned into the Amiga using a Sharp color scanner. The image was thenconverted from the Amiga's ILBM format to the run-length encoded (R LE) format used by Wavefrout, andthen converted into a texture map. The image was then mapped to a simple square and placed in front ofthe nossle model. The nozzle model was generated completely in Wavefront using the actual dimensions.The "camera" was positioned so that the drawing completely obstructed the nozzle from view. The drawingwas then animated moving back, passing through the model, and at the same time the camera was rotated.The length of the animation was 100 frames.

Spjnning NozzleThe model of the nozzle was modified to make a cut-away view of the interior, and was then animated

rotating 360' about all three axes in 100 frames. The final animation was too large to fit on the Amiga, sothe animation was recompiled using all odd-numbered frames, which is why the movement was jumpy.

Iaiection-molding SimulationiThe model of the mold tool was based entirely on conversations with Mr. Chris Frank, Mr. Rich Griffin,

and hand-drawn sketches of part of the mold. The nozzle model was used here as well.

Animation Procedures

All of the models were generated using the MODEL module of Wavefront's Advanced Visualizer. Ma-terials and texture mans are created using Wavefront's MEDIT (Material D[)l'ror) and then applied to theobject in MODEL. Motz. paths are created in Wavefront's PV (PreViewer) and objects are then assignedto a motion path. Individual frames are rendered using Wavefront's IMAGE module with shadows, reflec-tions, and antialiasing turned off to speed up the rendering. After each frame is rendered, it is convertedfrom Wavefront's RLE format to a 24-bit Amiga ILBM format using a program developed by Mr. RussLeighton. The 24-bit image is copied to an Amiga 2000 and converted to a 4-bit compressed NTSC videoimage using commercial software. After all the images have been rendered, the animation is compiled usinga commercial program on the Amiga which employs a delta compression method. The animation is thenplayed back on the Amiga and each frame is decompressed to (in real time) to a true NTSC video image,'sing commercial hardware. The output of the decompression hardwiaT is connected to the video-in jackof a standard VCR and recorded. The background music was pl : -1 fit, i,. \,mig ;., Ith,'-. imnti',n wnsplaying, taking advantage of the Amiga's multitasking qwm.railiit-: .. Iit,. ;mdi.... mipw,,iI jacks ',f 4hO,Amiga were connected to the audio input jacks of the V('H.

Most of the process described above has been automafed usiuug 'iini- cripfs and prfgrnauus developed InMessrs. Russ Leighton and Tom Elkins.

312

For More Information1-DEAS

Structural Dynamics Research Corporation (SDRC)Software Products Marketing Division

2000 Eastman DriveMilford, OH 45150-2789(513) 576-2400They do offer discounts and special programs for Universities.

WavefrontWavefront Technologies530 East Montecito St., Suite 106Santa Barbara, CA 93103(805) 962-8117The software was about S 35,000, but they do offer less expensive products as well; in fact, their Personal

Visualiser comes with some unix workstations. They also offer a Data Visualizer which has some incrediblecapabilities for viewing complex datasets. One may also see the capabilities of the Advanced Visualiser bywatching TV and movies such as Star Trek: The Next Generation, Lifesaver commercials, "Total Recall"(the X-ray screen), and many TV lead-ins to news, sporting events, and movies.

Compressed NTSC Video Images

The product is called DCTV (Digital Composite TeleVision)Digital Creations2865 Sunrise Blvd., Suite 103Rancho Cordova, CA 95742(916) 344-4825

3

313

Injection Molded Rocket Components

Christopher L. Frank

Advanced Composites Program Office -

Sacramento Air Logistics CommandMcClellan Air Force Base, Sacramento CA. 95652

ABSRACT

In September 1989, an informational meeting was held atMcClellan AFB to discuss the Advanced Polymer Components (APC)project under the direction of Dr. John Rusek of the Air ForceAstronautics Laboratory (AFAL) with the Advanced Composites ProgramOffice (ACPO). A co-operative effort began between the AFAL and theACPO to rapidly build a number of rocket motor and rocket engine partsusing a new type of plastic, Liquid Crystal Polymers or LCPs. Plasticmaterial had not been used for these type of applications before and agood deal of information had to be generated. The AFAL wanted toquickly establish an Air Force-staffed plastic motor program and cameto the ACPO for the expertise nee.ded to design the molds and developthe processes to produce these rnotors. By May of 1990, the timetablewas set, and design and anal;sis had begun. Molds were built, and onAug 28, 1990, less than 6 months from concept, the first eleven injectionmolded plastic solid rock,.. motor cases were fired. Seven of these casessurvived the firings. The initial success of this project convinced theAFAL to continue working with the ACPO in this area. The use of plasticcase designs for •olid rocket motors will contribute greatly to theultimate goal ),a a low-cost lightweight interceptor. This paper willhighlight thf. work to date, present test data and process information.

314

Liquid Crystal Polymers (LCPs) have a number of intriguingproperties that could prove very beneficial to the field of rocketry. Themost significant of these are, high strength fiber formation (fibulation),resistance to extreme temperatures, impact tolerance, and ease ofmolding highly detailed parts.

Figure 1 shows a typical solid rocket motor schematic. The variousmechanical parts and solid fuel contribute to the total weight of themotor. If this total weight can be lowered, through the use of newengineering polymers like the LCPs, increased payloads, increased fuelcapacity, or smaller rocket sizes may be realized. Beyond decreasingrocket motor weight the LCPs may also lower manufacturing costs, asvarious parts may be more efficiently manufactured by the use ofinjection or compression molding. For the purpose of this paper we willbe primarily concerned with the motor case, though other componentsare currently under development including a low pressure nozzle.

PROPELLANTGRAIN

GRAIN PORT THRUST M 7NOR ASSEMBLYAREA SKIRT MTRASML

(GAS CAVITY) CASE

Figure 1Generic Solid Motor Design

The 2X4 motor for testing solid propellants was suggested as oneof the first prototypes. This case is used for testing the formulation andburn rates of solid rocket fuels. Currently the 2X4 motor cases are madeof D6AC steel and are individually machined. They are reusable butrequire though cleaning and inspection. The size of the case is 2" insidediameter and is 4" long, thus the name 2 by 4 motor.(Fig 2)

315

The manageable size of this case, the varied fuel types possible, and thefact that an instrumented test fixture and test program were already inplace for the 2X4 made it an ideal candidate to become a test bed forthese new materials.

2.00"2.50"

-Y'

0.25"

Fig.22X4 Test Motor Case

Fibulation is a well documented characteristic1 of LCPs and so it isone of the APC goals to exploit this attribute. The use the injectionmolding grades of materials allowed us to capitalize on the naturaltendencies of polymer alignment. Injection molding induces mostplastics to flow in manner that causes the polymers to align near thewall where drag is high which induces shear. In the same flow thepolymer in and near the center of the flow is random and much lessaligned. The polymer associated with the wall forms a oriented region ofstrength in the direction of the flow, a sort of skin. The polymer in andnear the center remains in it's random state and forms a sort of core(Figs 3&4).

ILiquid Crystallinc PolymcrsNation Materials Advisory BoardCommission on cnginccring and Tcchnicai SyslemsNational Rescarch Council NMAB-453 1990

316

SKIN

CORE

UTZFECT

Fig. 3Typical Section View of Flow

The effect of the flow through a passage causes the alignment ofthe polymer. The result is that the wall becomes stronger due to thealignment but ordinarily only in the direction of the flow. The flow inthese molded rocket motor cases was all longitudinal so we can expectthe circumferential strength to be lower than that over the length, andthat is what was observed during initial testing.

SKIN

ORIGIN OFFLOW

COREEFFECT

FLOWFRONT

Figure 4Section of Flow

317

This skin and core effect enhances the fibrous formation of theLCPs. In figures 5a and 5b, the skin and core effect is quit visible andcan often be seen with only minor polishing.

41V V

'. 9i j, / .

* ' rV#

a bW

Figure 5Photo of Vectra A-950 Case

The skin and core effect can actually become so pronounced thatduring cooling of the molded article or under small loads the core pullsaway from the skin due to the large differences in strength associatedwith alignment. In figures 6a and 6b, the failures between the skin andcore are also quit visible.

318

• Je

a b

Figure 6Photo of Xydar SRT-300

Initial testing of the cases consisted of loading the test cases withsolid fuel and firing them in the instrumented test fixture. Once the testswere complete and results proved favorable the next design phase

began. This next step was to design a simple flight article and test it. Theflight article would be a motor known as the academy motor. This motoris a fairly simple design the current motor is made of a paper phenolictube that is plugged with a wood blank at one end and has a graphitenozzle pinned in the other end. To redesign this motor posed only twodesign problems, first, end containment and second, attachment of thenozzle. The containment was accomplished by designing a domed end(Fig. 7).

Fairly low pressures are involved, approximately 700 psi in thechamber at maximum thrust. The resulting load on the nozzleattachment is about 2000 pounds. The nozzle attachment required sometesting to determine the proper pin diameter to allow the plastic to carrythe load. This test was accomplished with a simple test fixture (Fig. 8)that could be pinned into the cases. The results of these test are thefollowing table.

I319

17/821/

0 91 /8 Do.. I II

Figure 7

Academy Motor

I LOAD

TEST MANDRIL

PINS

CASE

Figure 8Test Fixture for Pinning

As the test showed the larger diameter pins have been chosen forthe design. Although the pinned values are very close to the load values

320

they are considered to be with acceptable limits due to the use of anp additional adhesive to attach the nozzle (Fig.9).

An interesting obscrvation about the failures is that the rivetsbegan to bend prior to failure of the LCP. Photos of these failures werenot available at press but will be presented at the ThermoplasticReview. Also based on the type of failure the stronger LCPs exhibitedand the bearing failure of the rivets further work with a new test fixturewill be conducted.

Figure 9

Academy Motor Nozzle

During a recent end burn test a plastic motor case was exposed to a5000 degree flame at 50 psi for over 30 seconds. This type of conditioncan cut plate steel when using an oxy-acetylene torch.(See Videopresentation) As more applications are attempted and more testing ofthis material is accomplished the total potential of LCPs may soon berealized.

S321

Material Number Rivct Distance of Maxof Rivets Dia. Hole From End Load

VectraC-130 4 0.125 0.3125 1007 Lbs.Vectra 0.5625 &C-130 8 0.125 0.3125 2300 Lbs.VectraA-625 4 0.125 0.3125 1260 Lbs.VectraA-625 4 0.125 0.5625 1304 Lbs.VcctraA-625 4 0.125 0.5625 1435 Lbs.Vectra 0.5625 &A-625 8 0.125 0.3125 2615 Lbs.VectraA-950 4 0.125 0.5625 1470 Lbs.VcctraA-950 4 0.125 0.5625 1600 Lbs.VectraA-950 4 0.1875 0.5625 2 100 Lbs.VcctraA-950 4 0.1875 0.5625 2050 Lbs.VcctraB-950 4 0.125 0.5625 1850 Lbs.VectraB-950 4 0.125 0.5625 1400 Lbs.VectraB-950 4 0.1875 0.5625 2070 Lbs.VectraB-950 4 0.1875 0.5625 2250 Lbs.

XYDARSRT-300 4 0.125 0.5625 1150 Lbs.XYDAR

SRT-300 4 0.1875 0.5625 1240 Lbs.XYDAR

SRT-300 4 0.1875 0.5625 1350 Lbs.XYDAR

SRT-500 4 0.1875 0.5625 1290 Lbs.XYDAR

SRT-500 4 0.125 0.5625 2244 Lbs.XYDAR

SRT-500 4 0.1875 0.5625 1290 Lbs.XYDARRC-210 4 0.1875 0.5625 1500 Lbs.XY-ARRC-210 4 0.1875 0.5625 1627 Lbs.XYDARRC-210 4 0.1875 0.5625 1490 Lbs.

322

Air~

00

o o'

1Af

pil

U P 6

'- 'V

/f6

323~

.00

*A1

324

University of Lowell, Department of Plastics Engineering

S The work being performed by Dr. Nick Schott and Dr. Rob Nunn at theUniversity of Lowell, MA falls under the Macroscopic Material Propertiestask. Lowell has begun a research effort to injection mold test specimensand perform mechnical properties tests. Drs Schott and Nunn will developrecommended injection molding processing parameters for several LiquidCrystal Polymers. Room temperature mechnaical properties tests will beconducted to establish baselines for the material property database. Themechanical properties test will determine longitudinal tensile properties,Isotropic compressive properties, and longitudinal flexural modulus.

325

"Propella, Compatibility and Low Temperature CVD"

Dr. E.J. WuchererUDRI, OLAC PL/RKFC

(805) 275-5759, [email protected]

"Monomethylhydrazine (MMH) and Nitrogen Tetroxide(NTO) Compatibility with Liquid Crystal Polymers"

Thomas R. HillRKLC

APC Symposium21 February 1992

ABSTRACT: Tests conducted by Tom Hill indicated that many aromatic polyester basedLPCs were incompatible with Monomethylhydrazine (MMH) and Nitrogen Tetroxide(NTO). The LCPs generally gained weight when exposed to NTO for 24h at RT,consistent with some sort of radical aromatic nitration process. Similar exposure toMMH resulted in severe sample degradation, presumably due to the formation of ahydrazide and resultant rupture of the polymer backbone.

Chemical Vapor Deposition has been used to coat some LCP samples with Aluminumfilms at 110C. Careful control of the sample preparation, surface pretreatment anddeposition times are used to control the morphology of the metal film. Al coated Vectrasubstrates are resistant to attack by MMH, though film quality must be carefullycontrolled since any slight film defect can open a pathway for failure and substratedegradation. Several LCPs (esp HX-4000) are difficult to coat and may requiremechanical roughening of their "skins" before they can be successfully coated.

326

B

one

0it32

"a-(I,€I-• g" 82 / ' ' gO 0

U=~gW•E02

I0z0

CC

x0

S6~ Z~ £Z S9~N~gwosD

328

I-W

Um

sla a9904ISOO KDOI 1.1O' 000'1

329

L&JU.-

0N

L&J

9601, 0600 90 2eO a~ elO' 000',

330

IW>

m0 0

~ C .S

w0 0

W0o~___ C

__ + >4)

>1. I (U4 9 ~ M

-. 0

-~ cr (U.

=C-5 M)(

(ml) (Ua)N NCC

bmOMIN N

331

ZZ

VZZZZ Z V~

0 "t -, WI- C, '

0 +0

.2 -t

z 0

0 0- r-:

"o-

_ ~CA

E 0332l cl + c

Low Temperature CVD

What: Aluminum, nickel, copper or rhenium thin films.

Where: On polymer, liquid crystal polymer, orpolymer composite substrates.

Why: Light weight, inexpensive, oxidation resistant,propellant compatible metals with knownCVD processes.

Application: Tanks, tubes, valves or structural materials(0 atom resist).

3333

.. J

Aluminum Low Temp CVD

LiA1H4 + HN(CH 3)3C1 - AlH3*N(CH3)3 + LiCI + H2

I1 bCA1H3'N(CH3)3 Al(s) + 3/2H2 + N(CH 3)3

TM5x 10-5torr TiC14 A1H 3'N(CH3)3VectraTOMI Al-Vectra TM

30s 30min

334

Micro Tubes

Al CVD Coating

Graphite or Al CoatedPolymer Fiber Graphite or

Polymer Fiber

Polymer Fiber Carbon FiberSolvent OxidationRT 600C- Disolved Polymer -C02

I

AI Tube A1203 Tube

335

Al Coating Characteristics

1. Nucleation and initial growth has a very fine grainresulting in a highly reflective surface.

2. Growth proceeds on preferred crystalline faces,resulting in large (-1~g) grains.

3. "Cobble stone" growth may not be the best sealant.

0336

* "Dunk" Testing

Substrate: VectraTM liquid crystal polymer.

Coating: Aluminum, 4 x It.

Sample: Glass fiber filled injection molded cylindercut into pieces approx. 1.5" x 0.38" x 0.19".

Propellant: Monomethylhydrazine (MMH), propellantgrade.

Test: Samples immersed in MMH for 24h at RT,

Result: Uncoated sample weight loss > 50%.Coated sample weight change ±2%.

337

1 4 4

0-11

�1

�

� '4

I.

N.-

S PI

V I - 4A

a -C.. 4

rw

A '*S M-t *.~ .A

4.11

Conclusions

1. LCPs are significantly attacked by propellants

2. CVD coatings can reduce (prevent?) degradation.

Current Problems

1. Can CVD coating prevent degradation?

2. Will coated components survive long term application instressful environments?

342

2 X4 MOTOR DEMONSTRATIONS

TASK 19

The objective of this task is to determine if advanced poly-

mers, specifically liquid crystal polymers, can be used as solid

rocket case/insulation material. The tests performed in this task

include solid rocket propellant to LCP bonding, LCP motor case

strength, and the ablative properties of LCP's due to solid pro-

pellant exhaust.

The primary apparatus used for these experiments is a 2 in.

diameter by 4 in. length cylindrical rocket motor case.

Experiments done to determine if solid rocket propellant can be

bonded to LCP cases showed that typical solid rocket propellants

can be bonded to LCP materials with standard adhesives used in

conventional rocket motor cases. The strength of LCP motor cases

were determined by firing the motors at increasingly high pres-

sures until the cases fail. Ablation properties of LCP's were

studied using long duration (- 14 seconds) motor firings where a

section of each motor case is directly exposed to the high tem-

perature (> 2800 K) solid propellant exhaust products.

343

0LL U)

00 z

00

< 0 Z

LLJ

Goc

~;0

wU.

344

0 -J 0C/) cc ) WD w 0

DW<

z0< Z QF-

Z5zZZ7,H <cJ

0-J -~cn Dd-)

H0 za w

L - mL L o C ') CO ' Lw C U L DZ Cf) CO w

z8J ý8 LU 0D Z L Z(n ) c-

Z . O-ý W 1 H C-~ 0 moo,)0HH

mm o<2xI

345 WF

LUJ

0wl CL LU

0 <</ m jU

I LL <

o- 0C LUJ

F-- <

0 U]

Cl)U LUL

/) C/) Cl) U) U

w LU w 0 LHz Z L-j Lz

i -j LUWO 0

WIa: WIJ 0 L a

0 00 f 0 0

346 0 C f

Ir,

__ -- O

'21347

z0

U.. 0

00o z0

LL m

z.)0: z00

H8

Cowwix

CftCo-Ia:0.

COa:

348

LCP 2X4 FIRINGS

FIRING CASE PEAK AVERAGE DURATION CASE/PROPELLANT COMMENTSNUMPER MATERIAL PRESSURE PRESSURE BOND PROMOTER

(PSI) (PSi) (SEC)

1 VECTRA C130 961 864 1.446 N-100

2 VECTRA C130 1278 .070 NONE FAILED ONIGNITION

3 VECTRA C130 1018 990 1.376 NONE

4 VECTRA C130 1303 .059 N-100 FAILED ONIGNITION

5 VECTRA A625 966 .058 N-100 FAILED ONIGNITION

6 VECTRA A625 1019 .807 N-1O0 FAILED@ +.80 SEC

7 VECTRA A625 862 818 1.436 NONE

8 VECTRA A625 913 876 1.419 NONE

9 RYTON 316 269 2.346 NONE

10 RYTON 753 727 1.578 N-1O0

11 RYION 745 713 1.605 NONE

S349

'I.., ....- I ,

I.I� w1> 1

I\ *�'l. I

\\ K\\A'� j- I:I >'

* - *1.1. - --

N, I:

pI I

* I

I' ii'I.1w

N.

N

0

350

Sc0

zo z

a:-

0

0 0Lj 0

0

0>

- <U) ww wC.)

z Lcrr

z0

351

00

z-

ODU

000a:

Cl)

a: 04

m

352

I Lw 00

LLIL<

CI Uz 0_:c F- C/ 0

LL ~ < (rCr) 0 C-0 5 0 CLU o0 <0 OH ww

rr 0 zc <(1)0 ~ C) 1Zn. 0r (0<D t

0 F-C/(.00< 0<

00 C) C %0la.0 amm( Hz HZn<

WW _j

<CO < -w F-w D(H<H HC _jLI Zo L L

Cl) Z<~ 0 z) L< CC- CWOz~ <4 ZmCc/) H0 oHn oC/

I CL

0j o0L>U 0 F

m 353C/

CD

Ul)z 0

0C5w < /O<C/)

0 CF-X

LLW

z w,

354

No-Deposit, No-Return Spacecraft

J. T. KareLawrence Livermore National Laboratory, USA

In ground-to-orbit laser propulsion, a ground-based laser suppliesenergy to a small rocket vehicle. The laser energy heats an inert propellant,which is exhausted to provide thrust. The rocket can be both simpler andhigher in performance than a conventional chemical rocket, and (with somedesigns) can be steered from the ground, by controlling the laser beam.However, to make economical use of the laser, a very large number ofpayloads must be launched -- up to 30,000 per year. The individual rocketvehicles must therefore be very inexpensive.

One laser propulsion technology, under development at LLNL, uses ahigh-performance heat exchanger to absorb the laser energy and transfer itto the liquid hydrogen propellant. HIX (Heat eXchanger) vehicles basicallyconsist of a liquid hydrogen tank, a flat metal heat exchanger, and one ormore nozzles. The hydrogen is pressure-fed from the tank, at a typicalpressure of 70 psi, and heated to 1000 to 2000 K in the heat exchanger. Itis exhausted at a typical pressure of 25-30 psi.

If such vehicles are to be both feasible and economical, low-cost,light-weight components are essential. LLNL is developing a low-costelectroplating technique to fabricate heat exchangers. Advanced polymers,with their high strength to weight ratios, good thermal performance, andlow fabrication cost, could be useful for other parts. In particular, blow-molding might allow cheap fabrication of meter-scale liquid hydrogen tankswith better performance than aluminum tanks. Other structuralcomponents, and perhaps even "hot" parts such as nozzle skirts, might bemade by injection molding. The very short working life (5 minutes) andlow operating stresses of these parts would allow operation at the very edgeof their mechanical and thermal limits. A potential future advantage wouldbe that such plastic parts could be recycled in space, either as thermoplasticsor as a simple stockpile of carbon and hydrogen.

A near-term demonstration of HX technology is planned, using a 50kW laser and a 10 cm diameter heat exchanger. This "bottle rocket" willuse liquid nitrogen propellant, and will need a 1 to 2 liter propellant tank.Fabrication of such a tank would be a good test of blow-molding techniquesfor advanced polymers.

S355

"0

CCIH

0 "0

V.. -

C/) C13 U-

0 =/ "-'. 424

V-4-

0o0 C)

0 UM C cl 4m '7

U u o F-4 uw r-.4 .

-~0- clc~ = ..

u U0c->

356o

0l

a)C1 0a

0.

ci) N

H 0

4 -4

- -- 4>.b cz

0 o 0~

35

00

C,.)

"0~

CInS C) v4-

o C13- mco"0

1C ) I.

W)* 4--4

"o 0

.- A

Hý1 4- 4- 44J

"0 u

358 s

U-4 -4

00

U-4 4

IM44

1-

z 4)4

H Cd~H '

359

4AE=0 0s2C

~m

C6 M

"0 ~~C.44 )- 06

= .0

_ C.' 4 c

4-J ;.-4

I=L4 0

CI) 1-4=

z zC

360

0

0

- 0

z -

E

(D 0=LE >

o "0_C- 0C C-

14

Cu YUM

CDo 0 a 0 0 a 0 0 D 0a CD 0 CD a D a D a

o o 0D 0n 0 C0) 0 0 0 0

S 'dsl

3361

0 04l

owI 0 -r

E4

*4eC)

CL

U) 0

wE CL

E 0

U)E CDU)

to 4 0

Co ) I)

0

E

-00

(4( I I I I0)DO 0)c - (D to) W* m CJ 4 0

,OMWd/B)I

362

(A

0

0 E4.0cc CD

cc

0

CDC

00

3630

EEa E

a CI CD0 CD

mc 4.dE Cx 0)

VCt 0)C0

.. .. .. .. .. ..... ...V.. .. .. ... .. ... .

CcO.. .. ..... .......... ....

... . .. .. ... ... ... ... .. ....... ... . ... ..... .. .. ....

x~.... .. . .... ... . ..

...... ...... ...

........ _.. 0............ .... .... . .. .... . .... .... .... ..... . . . . . ... . . . . .. . . . ........ (. ..................

... .. .. .. . .. .. 0.

.. ... ... ... C cf1T~f~ffrn irir .............. 0

0n 0................

.. ...... .0)~0O 007O

.o ........... ...O

.. .. ... . 4. .

3

365

0 Qz

o o 02

0Z-> -. 4

~E

C))0q CZ 0>

C/ >

oo'Sl

Z 0

M00

361

iHiR 03 '92 14:13

DEPARTMENT OF THE AIR FORCEHEADQUARTERS OGDEN AIR LOGISTICS CENTER IAFLC)

HILL AIR F%)RCE BASE. UTAH 84050"5069

"w,1.v0o,.: OO-ALC/TIELM 3 Mar 92

Su"Cr. Interest in the APC Program

To AL/RKP (Dr Rusek/275-5430)

1. The purpose of this letter is to provide background informationon my interest in the APC program, provide any applicableinformation and capabilities, and define any future requirements.

2. My educational background is a B.S.in Engineering Science andMechanics. I am presently serving as an Air Force Captain at HillAFB. My present job specialty is working in a laboratoryenvironment performing failure analysis from a metallurgicalviewpoint. I have had training and experience in advancedcomposite structures and injection molding processing.

3. My interest in the APC program originated as a funds manager atHill AFB. My task was to program funds for injection moldingprojects and prototyping. My function is to act as a laisonbetween the plastics shop and the funding people to insure thematerials are available, work orders are accurate, and trainingfor the plastic shop personnel is available. Hill AFB is one oftwo sites within the Air Force that has injection moldingcapabilities. McClellan AFB has a large injection molding machinemanaged by Mr Chris Franks. Hill AFB has a smaller injectionmolding machine catable of molding parts up to approximatelytwenty ounces, managed by Mr Rich Griffin. Several of the partsthat were displayed at the recent APC Symposium were manufacturedat Hill AFB. Currently my interest lies in developing anunderstandinq Qf liUid cryst a l pnrym•s (LCPs) and tying that toan Air Force application. Current projects include molding athermoplastic radome for fighter aircraft.

4. Please call for specific information on the injection moldingcapabilities if you are interested or have a potential use. Inaddition, we have scanning electron microscopes (SEM) availablefor testing purposes in the metallurgica--l laboratory.

5. Future requirements will be to continue support of the APCprogram. Hill can terve as the distribution point of commercialgrade quantitias of LCPs for your research. Presently, we have onhand quantities of Zydar 300 & 500, HS4000, and Vectra.

368Sb -COMBAT STRENGTH THROUGH LOGISTICS

6. If any questions, please contact the undersigned commerciallyat (801) 777-2874 or 775-2482 and DSN 458-2874 or 924-2482. OurFAX number is (801) 777-8049.

STEVEN B. HARDY, CAPT, USAFMaterials and Processes EngineerMetallurgical SectionTechnology and Industrial Support

Directorate

S 369

1. P E. : . ! :07

ADVANCED POLYMER COMPONENTS RESEARCH SYMPOSIUM

Mercer Engineering Research Center1861 Watson Boulevard

Warner Robins, GA 31093

The primary focus on advancing the understanding and enhancingthe physical properties of Liquid Crystal Polymers (LCP) is thepursuance of applications in the astronautics community. LCPs areprime candidates for space applications because of their highstrength to weight ratio. However, the remarkable mechanicalproperties and thermal stability of LCPs also make them desirablecandidates for a multitude of more common applications where weightsavings are advantageous. These common applications range frommilitary aviation to the private automobile.

Mercer Engineering Research Center is currently interested inthe development of LCPs because of its applicability to militaryaviation. The design evolution of military and commercial aviationhas gone from primarily aluminum structures to a heavy reliance oncomposite materials. This reliance has been primarily fueled bythe increased strength to weight ratio of these compositematerials. However, composite materials are not without theirproblems. Delamination, loss of strength due to aging, andstructural integrity non-destructive testing are some of theproblems inherent with composite materials. As our military andcommercial aircraft design demands continually increase, newmaterials will be required to fulfill the dreams of these newdesigners. LCPs are prime candidates for these new materials.

While the aircraft industry is looking to the future andcontinually stretching material technology as it goes, there arestill hundreds of older aircraft that could also benefit from thisnew technology. Because of high replacement cost, these olderaircraft are being required to fly beyond their design life withadded demands of more aggressive mission profiles. As a result,structural failures are becoming common place. Instead of justrepairing the failures, upgrades to the design are generallywarranted. The designers of these upgrades must keep abreast ofnew technologies that may provide an alternative with enhancements.This has inherent benefits to the user,i.e. increased structuralintegrity, lower weight, and better performance.

As the technology of the LCPs matures, the areas of itsapplications will become better defined. As space is an obviousenvironment for LCPs, other more down-to-earth environments andapplications are logical extensions of this technology.

POC: Terry Fuchser, ph. (912) 929-6444

0370

Kevin E. MahaffyOLAC-Phillips LabRKCCEdwards AFB, Ca.93523-5000T(805)275-5450F(805)275-5144

RETICULATED BOND LINE RESEARCH

The goal of the Reticulated Bond Line project is to create a strong and

long lasting bond between the propellant and the insulation of the case wall. In

this concept, the propellant, with a polymer binder, flows into the cavities of the

open cell foam. The bond derives its strength from two sources. First of all, the

greatly increased surface area of open cell foam over which chemical bonding

can take place. Second, the propellant will polymerize around the ligaments of

the foam. The inter-penetrating networks will add mechanical strength to the

bond.

The primary technical problem is how to get a layer of open cell foam

joined to a layer of solid insulative material. The preferred solution would take a

material such as a Liquid Crystal Polymer(LCP) and form it into the shape of the

Missile case. The finished part would have a shell of solid LCP that transitioned

continuously into an open cell foam.

Any suggestion of a possible technical approach to this problem would be

welcomed.

371

p oc°

C0

0

373

10

t E UE(DLA. Eu

Qo

LLU

Lui

* a LUU

a o 0

o.......... a

C) i

374

proceedings annual advanced polymer dtic · advanced polymer components research symposium schedule...

Documents