AD-A246 346

AD

ORNUJTM-11759

OAK RIDGENATIONALLABORATORY

Characterization of Rocket- /A Propellant Combustion Products

Chemical Characterization andComputer Modeling of the Exhaust

,- Products from Four Propellant

DTIC Formulations

SEF-LECTEýFE 2 " 1 2R. A. Jenkins-U C.W. Nestor C. Y. Ma

C, V. Thompson B. A. TomkinsT. M. Gayle R. L. Moody

Thý(ijtriuuent h41 been aPprovedfor P~ll).HC IOU.'4ICtai rind ncile: its

MAG•Pa , 92-04722MARTIN MAIETTA ENERGY SYSTEMS, INC.FOR THE UNITED STATESDEPARTMENT OF ENERGY

This report has been reproduced directly from the best available copy,

Available to DOE and DOE contractors from the Office of Scientific and Techni-cal Information, P.O. Box 82, Onk lidge, TN 370311; priceo available from (61b)676-8401, FTS 626.8401,

Available to the public from the National Technical Information Service, US,Department of Commerce, 5285 Port Royal Rd,. Springfield, VA 22161,

This report was prepared es an account of work sponsored by an agency ofthe United States Government, Neither the United States Government nor anyagency thereof, nor any of their employees, makes any warranty, express orImplied, or assumes any legal liability or responsibility for the accuracy, corm-pleteness, or usefulness of any information, apparatus, product, or process dis-closed, or represents that Its use would not Infringe privately owned rights.Reference herein to any specific commercial product, process, or service bytrade name, trademark, manufacturer, or otherwise, does not necessarily consti-tute or Imply Its endorsement, recommendatlon, or favoring by the United StatesGovernment or any agency thereof, The views and opinions of authorsexpressed herein do not necessarily state or reflect those of the United StatesGovernment or nny agency threof.

DISCLAIMER NOTICE

THIS DOCUMENT IS BEST QUALITYPRACTICABLE. THE COPY FURNISHEDTO DTIC CONTAINED A SIGNIFICANTNUMBER OF PAGES WHICH DO NOTREPRODUCE LEGIBLY.

* ADORNL/TM- 11759

CHARACTERIZATION OF ROCKETPROPELLANT

COMBUSTION PRODUCTS

SUBTITLE:CHEMICAL CHARACTERIZATION AND COMPUTER

MODELING OF THE EXHAUST PRODUCTS FROMFOUR PROPELLANT FORMULATIONS

Final Report

DOE Interagency Agreement No, 1016-1844-AlProject Order No. 87PP8774

December 9,1991

Principal Investigator: R. A. JenkinsPrimary Contributors: C. W. Nestor, C. V. Thompson,

T. M. Gayle, C. Y. Ma, B. A. Tomkins, and R. L. Moody

Accusiron For Analytical Chemistry DivisionOak Ridge National LaboratoryDiIC tA[J P. O, Box 2008

Uwolutioced Oak Ridge, Tennessee 37831-6120disihficdtkio (615) 576-8594

a........-

y .. .y .............. .......Di t iulu) IOI

... ....... ~.*.- ...... Ms. Karen FritzAv, iI,,/ L'.., Chief, Acquisition Management Liaison Office

U.S. Army Biomedical Research and DevelopmentDi&A Laboratory, Fort Detrick,

Frederick Maryland 21701-5010

COR: Major John Young

* ('i,•, *] •

RECRT DCC-; % - - AJ' GE "MG9 EC

:7 7 .7 W '.m llu

1. AGENCY USE ONLY (Luae'le oarV' 2.P OORT 3ATE .. AEORT .P•E AND DATES .OVERED!December 31, 1990 Final Report, 9/23/87 - 4/1/90

4, TITLE AND SUBTITLE CHARACTERIZATION OF ROCKET PROPELLANT COM- I S. FUNDING NUMBERSBUSTION PRODUCTSSubtitle: Chemical Characterization and Computer MOdeling APO 87PP8774of the Exhaust Products from Four Propellant Formulations PE - 62720A

6, AUTHOR(S) . PR - 3M162720A835R. A. Jenkins, C. W. Nestor, C. V. Thompson, T. M. Gayle, TA - 00,C. Y.Ma, B. A. Tomkins, R. L. Moody WUDA314033

7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8, PERFORMING ORGANIZATIONREPORT NUMBER

U.S. Deparment of Energy ORNL/TM-11759Oak Ridge Operations OfficeP.O. Box 2001 DOE IA No. 1016-1844-AlOak Ridge, Tennessee 37831-8622

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

U.S. Army Medical Research and Development Cormmand AGENCY rEPORT NUMBERFort Detrick, Frederick, Maryland 21702-5012

U.S. Army Biomedical Research and Development LaboratoryFort Detrick, Frederick, Maryland 21702-5012

II. SUPPLEMENTARY NOTES

12s. DISTRIBUTION/ AVAILABILITY STATEMENT 12b, DISTRIBUTION CODE

Approved for public release; distribution unlimited

13, ABSTRACT ,Mxmum 200 words)

The objective of this work was the determination of the chemical composition of exhaust products forom thefiring of scaled down rocket motors at the Army Signature Characterization Facility at Redstone Arsenal, andthe comparison of those results with component levels predicted by a selected computer model. Both real timeand off-line sampling and analysis approaches were employed, Four types of propellant compositions wereevaluated. CO levels ranged from 85 350 ppm, while particle concentrations ranged from 30 - 100 mg/mn.All of the airborne particles were in the inhalable range. For two of the propellants, airborne lead was greaterthan 10 mg/m3. For the predominantly perchloiate formulation, hydrogen chloride (HCI) levels were greaterthan 100 ppm. Particulate PAHI levels were about a factor of 10 lower than that in outside ambient airparticulate matter. The computer model predicted mole fractions for CO were typically 20 35%, except forthe predominantly inorganic formulation, The model correctly predicted only minor amounts of ammonia andessentinlly no hydrogen cyanide. The accuracy of the predicted CO/CO2 ratios was low for all but one of theformulations. A modification of the model accomplished by mathematically accounting for minxing of hot exhaustgases with ambient air brought the predicted CO/CO2 r'ito Into greater agreement with that which was observedexperimentally.

14, SUBJECT TERMS 15. NUMBER OF PAGESPropellants; Chemilcal Characterization; Computer Modeling ofCombustion Productb; RA I111 PO 16. PRICE CODE

17, SECURITY CLASSIFICATION I8. SECURITY CLASSIFICATION 19. SECURITY CLASSIFICATION 20. LIMITATION OF ABSTRACTOF REPORTI OF THIS PAGE OF ABSTRACT

Unclassified Unclassified Unclassified Unlimited

NSN 7540-01.•.8t5500 stancarcd form 198 ,Rov .>.c9)a.,v om, Ov ANSI¢% sI• ,j '

, 3 2 j,

6

GENERAL INSTRUCTIONS FOR COMPLETING SF 298The Report Documentation Page (RDP) is used In announcing and cataloging reoprts. It Is Importantthat this information be consistent with the rest of the report, particularly the cover and title page.Instructions for filling in each block of the form follow. It Is Important to stay whthin the lines to meetoptical scanning requirements. C

Block 1. Agency Use Only (Leave blank). Block 12a. Distribution/Availability Statement.

Block 2. eport Date. Full publication date ")enotes public availability or limitations. Cite any

Incuding day, montha nd year, if available (e.g. 1 availability to tho public. Enter additional

ian 68) Must alinbion or special markings In all capitals (e.g.

Block 3. Type of Report and Dates Covered. 6

State whether report Is Interim, final, etc. If DOD - See DoDD 5230.24, "Distributionapplicabl, enter inclusive report dates (e.g. 10 Statements on TechnicalJun 87 - 30 Jun 86). Documents."

Block 4. Title and Subtitle. A title is taken from DOE - See authorities.

the part of the report that provides the most NASA - See Handbook NHB 2200.2.

meaningful and complete Information. When a NTIS - Leave blank.

report Is prepared in more than one volume,repeat %he primary title, add volume number, and Block 12b. Distribution Code.include subtitle for the specific volume. Onclassified documents enter the title classification DOD - Leave blank.in parntheses. DOE - Enter DOE distribution categories

Block 5. Funding Numbers. To Include contract from the Standard Distribution forand grant numbers; may include program Unclassified Scientific and Technical

element numbers(s), project number(s), task Reports.number(s), and work unit number(s). Use the NASA - Leave blank.following labels: NTIS - Leave blank.

C - Contract PR - Project Block 13. Abstract. Include a brief (MaximumPE - Program WU - Work Unit 200 words) factual summary of the most

EPement Accession No. significant Information contained In the report.

Block 6. Author(s). Name(s) of person(s)responsible for writing the report, performing Blocid 14. Subject Terms. Keywords or phrasesthe research, or credited with the content of the identifying major subjects in the report.report. If editor or compiler, this should followthe name(s). Block 15. Number of Pages. Enter the total

Block 7. Performing Organization Name(s) and number of pages.

Address(es). Self-explanatory. Block 16. Price Code. Enter appropriate price

Block 8. Performing Organization Report code (NTIS only).Number. Enter the unique alphanumeric reportnumber(s) assigned by the organization Blocks 17.-19. Security Classifications. Self-performing the repor. explanatory. Enter U.S. Security Classification in

Block 9. Sponsoring/Monitoring Agency Names(s) accordance with U.S. Security Regulations (i.e.,and Address(es). Self-explanatory UNCLASSIFIED). If form contins classifiedBlock 10. Sponsoring/Monitoring Agency information, stamp classification .i the top and

Report Number. (If known) bottom of the page.

Block 11. Supplementary Notes. Enter Block 20. Limitation of Abstract. This block mustInformation not included elsewhere such as: be completed to assign a limitation to thePrepared In cooperation with...; Trans. of ... ; To be abstract. Enter either UL (unlimited) or SAR (samepublished In... When a report Is revised, Include as report). An entry In this block Is necessary Ifa statement whether the new report supersedes the abstract is to he limited. If blank, the abstractor supplements the older report. Is assumed to be unlimited.

Standard Form 21M Back (Rv

AD _ORNL/TM- 11759

ARMY PROJECT ORDER NO: 87PP8774

DOE Interagency Agreement No, 1016-1844-Al

TITLE: CHARACTERIZATION OF ROCKET PROPELLANT COMBUSTIONPRODUCTS

SUBTITLE: CHEMICAL CHARACTERIZATION AND COMPUTER MODELINGOF THE EXHAUST PRODUCTS FROM FOUR PROPELLANTFORMULATIONS

PRINCIPAL INVESTIGATOR: R. A. JenkinsPRIMARY CONTRIBUTORS: C. W. Nestor, C. V. Thompson, T. M, Gayle

C, Y. Ma, B, A, Tomkins, R, L, Moody

CONTRACTING ORGANIZATION: U.S. Department of EnergyOak Ridge Operations OfficeP. 0, Box 2001Oak Ridge, Tennessee 37831-8622

REPORTED DATE: December 9,1991

TYPE OF REPORT: Final Report

SUPPORTED BY: US.ARMY BIOMEDICAL RESEARCH AND DEVELOPMENTCOMMANDFort Detrick, Frederick, Maryland 21701-5010

PREPARED FOR: Contracting Officer's RepresentativeU.S. Army Biomedical Research and Development LaboratoryFort Detrick, Frederick, Maryland 21702-5010Major John Young, Contracting Officer's Representative

DISTRIBUTION STATEMENT: Approved for public release: distribution unlimited

The findings in this report are not to be construed as an official Department of the Armyposition unless so designated by other authorized document.

Page intentionally left blank.

0 FORWARD

Opinions, interpretations, conclusions and recommendations are those of the author andare not necessarily endorsed by the U.S. Army.

.. Where copyrighted material is quoted, permission has been obtained to use suchmaterial.

Where material from documents designated for limited distribution is quoted,permission has been obtained to use the material.

-• Citations of commercial organizations and trade names in this report do notconstitute an official Department of the Army endorsement or approval of theproducts or services of these organizations.

/ In conducting research using animals, the Investigator(s) adhered to the"Guide for the Care and Use of Laboratory Animals," prepared by theCommittee on Care and Use of Laboratory Animals of the Institute ofLaboratory Animal Resources, National Research Council (NIHPublication No, 86-23, Revised 1985).

N/A For the protection of human subjects, the investigator(s) have adhered topolicies of applicable Federal Law 45CFR46.

1 ture Date

P

Page intentionally left blank.

S2

EXECUTIVE SUMMARY

The overall objective of the work described in this report is four-fold: to a) develop astandardized and experimentally validated approach to the sampling and chemical andphysical characterization of the exhaust products of scaled-down rocket launch motors firedunder experimentally controlled conditions at the Army's Signature CharacterizationFacility (ASCF) at Redstone Arsenal in Huntsville, Alabama; b) determine thecomposition of the exhaust products; c) assess the accuracy of a selected existing computermodel for predicting the composition of major and minor chemical species; d) recommendalterations to both the sampling and analysis strategy and the computer rmodel In order toachieve greater congruence between chemical measurements and computer prediction.

Analytical validation studies were conducted in small chambers at the Oak Ridge NationalLaboratory (ORNL), while the actual firings were conducted at Redstone Arsenal. Realtime determination of selected species was performed by a variety of techniques, includingnon-dispersive Infrared spectrometry, chemiluminescence, electrochemical monitoring, andoptical scattering. Samples for analyses of trace constituents were collected fromindividual firings In the ASCF, and returned to ORNL for analysis, usually by gaschromatography/mass spectrometry, Four types of propellants were examined: a doublebase, a double base with 8% potassium perchlorate, one propellant which waspredominantly ammonium perchlorate, and a minimum signature reduced smokepropellant, which was about two-thirds octahydro-l,3,5,7-tetranitro-1,3,5,7-tetrazocine(HMX). Small, 2x2 motors, containing 25 - 75 g of propellant, produced significantquantities of carbon monoxide (CO) and particles when fired into the 20 m3 chamber. COlevels ranged from 85 - 350 ppm. This is equivalent to reaching 2500 - 7500 ppm If a fullscale motor was fired In a similarly sized enclosed environment. Particle concentrationsranged from 30 - 100 mg/m 3. All of the airborne particles were in the inhalable range.For two of the propellants ( the double base and the minimum signature), airborne leadwas greater than 10 mg/m3 . No ammonia or hydrogen cyanide was detected above I ppm.For the predominantly perchlorate formulation, hydrogen chloride (HCI) levels weregreater than 100 ppm in the ASCF chamber. Because of the relatively high backgroundlevels observed, trace organic vapor phase constituents were difficult to accurately quantify.While a wide variety of trace constituents were observed, only a few were present at levelsgreater than a few ppbv. Compounds present at levels greater than 10 gsg/m 3 includedbenzene, methyl crotonate, toluene, and cyanohenzene. A number of PA-s and nitro-fluorene were observed in the airborne particulate matter. However, the levels were abouta factor of 10 lower than that in outside ambient air particulate matter at a militaryinstallation.

Computer modeling was performed with the NASA-Lewis CET-86 version. This approachobtains estimates of equilibrium concentrations by minimizing free energy. Mole fractionsof major and milnor species were estimated for a range of exit/throat area ratios. Thepredicted mole fractions for CO were typically 20 - 35%, except for the predominantlyinorganic formulation. The model correctly predicted only minor amounts of ammonia

3

and essentially no hydrogen cyanide. Predicted mole fractions did not vary a great dealwith such input parameters as exit/throat area ratios or small changes in the heats offormation of the various compositions, The accuracy of the predicted CO/CO2 ratios waslow for all but one of dhe formulations, In general, if the model were to be used in itspresent state for health risk assessments, it would be likely to over-estimate exposure toCO.

Probably the greatest limitation of the model is its inability to account for reactions afterhot exhaust gases leave the rocket motor nozzle. For example, the model predicted nosignificant quantities of NO would be produced, yet such was measured at ppm levels onevery burn. A modification of the model accomplished by mathematically accounting formixing of hot exhaust gases with ambient air brought the predicted CO/CO2 ratio Intogreater agreement with that which was observed experimentally, It seems likely that withthe appropriate modifications to account for the roles of kinetically governed processesand the afterburning of exhaust gases, the model could make a more accurate predictionof the amounts of the major products. However, it seems unlikely for the system to bemodifiable to the extent to which accurate predictions of toxic or carcinogenic speciespresent at the ppbv level could be made.

4

0 TABLE OF CONTENTS

Forw ard ........................................................... 1

Executive Sum m ary .................................................. 3

Table of Contents ................................................... 5

List of T ables ...................................................... 7

List of Figures ..................................................... 11

Acknowledgement .................................................. 12

I. Objectives .............................................. 13

II. Background ................................................. 13

Part 1: Chemical Characterization Studies

Experim ental ................................................ 15

Results and D iscussion ......................................... 15

SSummary and Recommendations - Part 1 ................................. 35

Part 2: Modeling for Health Hazard Prediction

Introduction ................................................. 37

Results and D iscussion ......................................... 39

Limitations and M odifications .................................... 46

Recommendations for Futher Work ............................ 52

References ....................................................... 53

Appendix A ....................................................... 56

Seleted Rocket Propellant Formulations

Appendix B .................................................. 61

Trace Organic Vapor Phase Constituents Observed In Selected RocketExhaust Atmospheres

O5

TABI.E OF CONTENTS (Cont'd)

A ppendix C ....................................................... 69

Output from Selcted Runs of Computer Model NASA-Lewis CET-86

D istribution List ............................................. ... 121

6

Table LIST OF TABLES

1 Summary of Sampling and Analysis Strategy for RocketExhaust Constituents at ASCF ............................ ...... 17

2 Summary of Characteiization Data Composition DM ajor Constituents ............................................ 20

3 Summary of Characterization Data Composition HM ajor Constituents ............................................ 21

4 Summary of Characterization Data Composition LM ajor Constituents ............................................ 22

5 Summary of Characterization Data Composition QM ajor Constituents ............................................ 23

6 Mean Concentrations Achieved in ASCF Chamber .................... 24

7 Particle Size Distribution Rocket Exhaust Particulate MatterM ean V alues ................................................ 26

8 Concentration of Selected Constituents in Chamber Blanks ............. 27

9 Estimated Concentration of Trace Vapor Phase ConstituentsComposition D ............................................... 29

10 Estimated Concentration of Trace Vapor Phase ConstituentsComposition H ............................................... 30

11 Estimated Concentration of Trace Vapor Phase ConstituentsCom position L ............................................... 31

12 Estimated Concentration of Trace Vapor Phase ConstituentsComposition Q ............................................... 31

13 Non-Siloxane Vapor Phase Compounds Present in Motor Exhaustsat Concentrations Greater Than 10 pig/m 3 in ASCF Chamber ............ 32

14 Concentrations (pg/g) of Nitro-PAH and PAH in ParticulateMatter Collected on Course Filters at ASCF Compared withOutdoor Air Particulate at US Army Installation ..................... 34

15 Exit/Throat Area Ratio Ranges Test Motor Configurations ............. 38

16 Predicted Mole Fractions as a Function of Exit/Throat Aia RatiosCom position D ............................................... 40

7

LIST OF TABLES (Cont'd)Table Pa_.g17 Predicted Mole Fractions as a Function of Exit/Throat Area Ratios

Composition H ............................................... 41

18 Predicted Mole Fractions as a Function of ExitThroat Area RatiosCom position L ............................................... 42

19 Predicted Mole Fractions as a Function of Exit/Throat Area RatiosCom position Q ............................................... 43

20 Effect of ± 5% Shift in Heat of Formation of Ammonium PerchlorateComposition L - Predicted Mole Fractions .......................... 45

21 Comparison of Observed and Predicted Carbon Monoxide:Carbon Dioxide Ratios ......................................... 46

22 Comparison of Observed and Predicted Concentrations of ExhaustConstituents in ASCF Chamber .................................. 49

23 Effect of Choice Gaseous Equation of State on Computed Mole Fractions forComposition H ............................................... 50

24 Influence of Exhaust Gas Mixing with Air on Carbon

Monoxide/Carbon Dioxide Ratios. Composition D .................... 51

A-1 Composition "D" Formulation ................................ 57

A-2 Composition "H" Formulation ................................. 58

A-3 Composition "L" Formulation ................................. 59

A-4 Propellant "Q" Formulation ..................................... 60

B-i Compositions "D and H" Concentrations ........................... 62

B-2 Composition "L" Concentrations ............................... 67

B-3 Composition "Q" Concentrations ................................. 68

C-1 Composition "D" Output ....................................... 70

C-2 Composition "H" Output ....................................... 85

8O

LIST OF TABLES (Conm'd)

-pU.e

C-3 Composition "L" Output ...................................... 100

C-4 Composition VQ Output.................................................... 112

9

Page intentionally left blank.

I0

LIST OF FIGURES

1 Time Course of Exhaust Products Post Firing. Composition D .......... 18

2 Time Course of Exhaust Products Post Firing. Composition H .......... 19

01

ACKNOWLEDGEMENT

We wish to thank the following individuals for their assistance with thisproject:

Ms. B. J. McBride, of the NASA-Lewis Research Center, for provision ofthe computer model used In the project; Dr, Eli Freedman, for assistancewith interpretation of the results of the computer modeling; Mr. L. B.Thorne and his staff, of the U.S. Army Redstone Arsenal, for theconstruction and firing of the 2x2 rocket motors, the provision of samplesof the various propellants, and the use of the Signature CharacterizationFacility and chamber; Dr. Steve Hoke, of the U.S. Army BiomedicalResearch and Development Laboratory, for the use of the on-line hydrogenchloride measurement system, and Major John Young, of the U.S. ArmyBiomedical Research and Development Laboratory, for his patience,support, and technical assistance in a number of the aspects of this project.

12

I. OBJECTIVES

The overall objective of the work described in this report is four-fold: to a) develop astandardized and experimentally validated approach to the sampling and chemical andphysical characterization of the exhaust products of scaled-down rocket launch motors firedander experimentally controlled conditions at the Army's Signature CharacterizationFacility (ASCF) at Redstone Arsenal in Huntsville, Alabama; b) determine thecomposition of the exhaust products; c) assess the accuracy of a selected existing computermodel for predicting the composition of major and minor chemical species; d) recommendalterations to both the sampling and analysis strategy and the computer model in order toachieve greater congruence between chemical moasurements and computer prediction,

II. BACKGROUND

Upon initiation of the Army's Health Hazard Assessment Program in 1983, the lack ofInformation on the potential health hazards from weapons combustion products, to includerockets and missiles, became evident, Research to elucidate significant health effects ofrocket and missile combustion products has been limited, Experiences with weaponssystems such as ROLAND, VIPER, HELLFIRE, STINGER, and MLRS have resulted inthe development of specific medical issues by the U.S. Army, Presumably, these issues willbe addressed, in order to enhance the effectiveness of soldiers using such weapons.Requisite to addressing these issues is defining the chemical and physical nature of thecombustion products.

Evaluation of rocket exhaust toxicity from Army missile and rocket systems has beendirected towards a limited number of combustion products. Chemical species such ascarbon monoxide, carbon dioxide, nitrogen, oxides of nitrogen, hydrogen chloride, sulfurdioxide, ammonla, lead, and copper are among those frequently evaluated, A USAMRDCstudy1 has demonstrated more than one hundred chemical species in the combustionproducts of selected propellants. Many of the species represent potential health hazardseven though the majority of those identified were at low levels, During the study, datawere obtained for the Multiple Launch Rocket System's (MLRS) propellant by computerprediction and laboratory analyses, The combustion product was generated by burning thepropellant in a small test motor. When the exhaust plume was vented into a chamber withan inert atmosphere, good quantitative data was obtained for twelve chemical species, andwas in excellent agreement with theoretically computed values. In excess of fifty trace gasspecies also were qualitatively identified,

Various investigators have examined propellant and related combustion products generatedin a variety of ways to include directly from a weapon or other equipment systeml' 5 ,burning in a calorimeter or bomb" 9 , personal and general area sampling in indoor firingranges°'0 11, and detonation or combustion in chambers or microcombustors 2'14"17. Themethods of sampling and characterizatlon also have been varied, Sampling has been doneunder atmospheric1',24.5,12, 16,and less than atmospheric 1-3'8'9' 13-15 conditions which providea basis for comparing the relation between variables, such as, pressure and available

13

oxygen, on the composition of the combustion product, Sampling methods have beeneither direct and continuous, e,g.,the method used by Goshgarlan1 3'14 where the exhaustproducts of solid propellants were introduced directly into a mass spectrometer for analysisimmediately following combustion, or by collection in a container or on a medium forsubsequent analysis. The latter has involved cryogenic trapping, evacuated glass orstainless steel cylinders, and sotbent cartridges, filters, and condensation trains, Analyticalmethods to detect organics, gases, metals, and particulates have included gaschromatography (OC), gas chromatography-mass spectroscopy (GC-MS), titration, opticaland infrared spectroscopy, scanning electron microscopy (SEM), x-ray emission anddiffraction, and particle size analysis. Because of limitations with each sampling andanalytical technique, several techniques must be employed simultaneously to optimizequalitative and quantitative characterization,

Computer models have been used to predict propellant ballistic properties to include theidentity of the major chemical species contained in the combustion products,' 3,s0 1 .When compared with laboratory derived empirical data, the models tend better to predictthi major species than the minor ones both qualitatively and quantitatively1 ,5 ' 19, Themodels predict the chemical species that occur at the nozzle of the rocket as the exhaustexits; however, afterburning changes the chemical content of the combustion product.Afterburning and Incomplete combustion effects are not predicted by the models,

The approach taken In this study was to carefully validate real time analytical methods inchamber studies at Oak Ridge National Laboratory (ORNL) for as many of the majorconstituents as practical, The instrumentation for real time monitoring would then betransported to the ASCF for the firing of the scaled-down test motors. Vapor and particlephase samples for determination of trace organics and metal species would be returnedfor analysis. The Army Signature Characterization Facility (ASCF) has been used todetermine the concentrations of major toxic species in propellant exhaust, e.g.,carbonmonoxide carbon dioxide, hydrogen chloride, lead, aluminum oxide, and other nuisanceparticles2d. The facility isa 19.6m 3 walk-in, climatic chamber with temperature limits of-40" to 140*F and humidity control in the range of 20 to 100% relative humidity (RH).Typical operating parameters are 70'F and 60% RH. Designed as a smoke measurementfacility, the ASCF has been adapted for the measurement of rocket motor signature andexhaust constituents. The facility serves as a large gas cell in which the exhausts ofstandard 2 x 2 motors can be measured by infrared spectroscopy (Fourier TransformInfrared Spectroscopy, FTIS), Ports in the ASCF allow sampling and measurement byother methods, e,g.,air sampling pumps and direct reading Instruments.

The results of the characterization studies were then to be compared with values predictedusing the most recent version of a computer model developed by the Lewis ResearchCenter of the National Aeronautics and Space Administration (NASA-Lewis). The modelwas then to be refined to the extent of available resources, In order to Improve thepredictive capability of the system.

Results of these studies are described in two parts. In Part 1, results of the chemical andphysical characterization studies are described and discussed. In Part 2, results of the

14

computer modeling work are described. Comparisons with characterization data are

performed, and recommendations for model improvement are made.

PART 1: CHEMICAL CHARACTERIZATIONSTUDIES

EXPERIMENTAL

The sampling and analysis methods used in this study have been described in detail in aprevious report21, and are summarized in Table 1, An assortment of real-time analyticalinstrumentation was employed, However, resources were not available for the use of on-line mass spectrometric measurement, as such would have required periodic tran.%port tothe ASCF. Essentially, the approach taken was to first validate candidate analyticalmethods in small chambers (0A4 and 1,4 mi3) at ORNL, Analytical measurements usingreal time instrumentation were made of target species in the presence of well definedquantities of other species. The extent to which these materials altered the response tothe target species was noted, and corrections made when appropriate. For species whichcould not be determined in real time (usually trace organic vapor phase and particle phasespecies), samples would be taken at the actual burns to be conducted at the ASCF, andreturned to ORNL for detailed chemical analysis. Following method validation for thepropellant composition of interest, the sampling and analysis instrumentation wastransported to the ASCF at Redstone Arsenal, and deployed for monitoring and sampling,Typically, between 2 and 3 firings of a test motor could be conducted during each 8-hourshift, Burns of the various propellant formulations took place between August, 1987 andDecember, 1989.

RESULTS AND DISCUSSION

The compositions of the various propellant formulations tested in this project are listedin Appendix A. Briefly, Composition D was a double-base propellant, comprised ofapproximately 50% nitrocellulose and about 40% nitroglycerine. Composition H was alsoa double base system, with approximately 8% by weight of potassium perchlorate added,Composition L was a formulation comprised of nearly 75% ammonium perchlorate, withthe remainder being polyvinylchloride plastic and di (2-ethylhexyl) adipate. CompositionQ was a minimum signature propellant, comprised of 66% HMX, and about 11 % each ofnitroglycerine and butane triol trinitrate, (A fifth motor, referred to as Composition Xwas fired only one time, and no modeling studies were applied to it.) (Note that thelinkage between the propellant and the weapon systems for which they may be used isconsidered CLASSIFIED information, Those having need of this information shouldcontact the COR listed on the title page of this document.) All of the propellantscontained small amounts of metals, The motor size tested varied between ca. 24 - 75 g,This compares to a typical launch motor weight on an anti-tank weapon system of ca, 560

8.

Sampling of the exhausts was not without its difficulties. For example, for the first run ofComposition D, the high volume particulate collector was placed inside the ASCF

15

chamber. However, the shock wave from the firing was sufficient to blow the filter mediaout of the holder, Thus, for subsequent runs, the sampler was placed outside the chamberand

0

16

TABLE 1Summary of Sampling and Analysis Strategyfor Rocket Exhaust Constituents at ASCF

C_.n.. nent Sampling and AnalYsis Metht-d

Carbon Monoxide Real Time, non-dispersive infrared analyzerCarbon Dioxide Real time, non-dispersive infrared analyzerOxides of Nitrogen Real time, chemiluminescence anaiyzerHydrogen Cyanide Real time, electrochemical analyzerAmmonia Real time, electrochemical analyzerHydrogen Chloride Real time, ion selective electrodeTotal Suspended Particulate Matter Real Time: forward scattering infraredphotometer

Off line: two-stage high volume filter,gravimetric analysis

Metals Low volume collection on membrane filter,followed by inductively coupled plasma oratomic absorption analysis,

Particle Size Distribution Cascade impaction, optical comparison ofstages

Trace Vapor Phase Organics Collection on multi-sorbent traps, followedby thermal des orption gaschromatography/mass spectrometric analysis.

Trace Particle Phase Organics Collection on twn-stage, high volume filter,analysis by .,igh perfotmance liquidchromna togra phy and/or gaichromatography/mass spectrometry.

17

connected to It with the flexible plastic pipe. Also, on a latter run with "D,"the force ofthe shock wave buckled the main chamber access door on the ASCF. For the final firingof "D,"the nozzle was changed to force the propellant to burn over a longer period oftime. This resulted in a considerable alteration in the exhaust composition (see Table 2).

Major Constituents

The observed exhaust major constituent concentrations in the ASCF are reported in Tables2 - 5, along with various physical characteristics of the motors, The data Is summarizedin Table 6,

It Is Important to note that for those constituents determined In real time (le, the gases),the concentrations listed represent peak concentrations, For gases, maxima were typicallyachieved within 30 seconds of the firing of the rocket motors. Presumably, maxima wereachieved as the chamber contents were mixed by the fan mounted inside the chamber.Such was not always the case for the particulate phase species, For example, In FiguresI and 2 are compared the time courses for some of the major exhaust products for firingsof Composition D and H motors, from about 30 seconds following the firing onward. ForComposition D, immediately after following the achievement of maximum concentrations,the constituent levels slowly decreased. While the same happened for Composition Hvapor phase species, the particles were very slow to reach a maximum, Although particle

go~ - ___

/00

BO

,,10

¢0

8 20

0 4 6 12 I

Time', minutes

U Partilces, ffgl m^ 3 + CO, ppV ILI 0 NO.~, ppmkI0

Figure 1. Time course of exhaust products post firing. Compositior D,

18

i illM

0

go

m 70

Sso

O 40

L30

20

10

S7 9 11 13 1s 17 19

Elapsed Tim41. minutela+ cc, ppr(v 10 0 NOX, ppm a ParticleOs, mf•lm3

0 FIgure 2. Time course of exhaust products post firing. Composition H,

size differences between the two products were minimal (see below), it was speculated thatthe action of the fans could have stirred up larger agglomerates which settled Immediatelyafter firing, which eventually broke up to form smaller primary particles. Cor-centrationreductions seemed most likely due to leaking of the ch:imber contents through door seals,bulkheads, etc. Particle concentrations decreased somewhat more rapidly than those ofvapor phase constituents, probably due to settling.

No attempt was made to determine the concentrations of methane, hydrogen gas, or watervapor. For the former two species, quantitative measurements would be very difficultwithout the use of an on-line mass spectrometer, and such was not available tir this work.Water vapor is one of the major components of the motor exhaust. The mole fractionpredicted by the NASA-Lewis computer program typically is in the range of 20% (seebelow). However, the difficulty of making accurate determinations of water vaporconcentration in a large chamber Is considerable. For example, the maximum amount ofhydrogen in any of the formulations listed in Tables A-i - A-4 is sufficient to produce only15 g of H20 in the 20 m3 ASCF chamber. This is comparable to increasing theconcentration by at most 0."5 g/m3, to a concentration of ca. 11 g/m3 at 60% relatlwihumidity at 210 C. The addition of this amount of water vapor would increase the RHby 4%, as long as no change in the temperature occurred. Giv 'hat such small changeswould be difficult to measure accurately, and that ,rater vapor rL,, s no health hazard,

0 19

TABLE 2

SUMMARY OF CHARACTERIZATION DATACOMPOSITION 0

MAJOR CONSTITUENTS

m0 omo m e=

RUN NUMBER 1 2 8 4 a ad

DATE 8-25-87 825-47 -264-87 6-26417 6-2348 6.23.88

OUANTI1Y OF PROPELLANT, g 75 71 75 75 67 NR

EXIT DIAMETER, Inches 4 1.0 1.0 1.0 1.0 1.0 1.0

THROAT DIAMETER Inohes 0,55 0.707 0.50 0,50 0.50 NR

ASCF CHAMBER TEMPERATURE, 'F 71 78 71 71 68 71

ASCF RELATIVE HUMIDITY, % 76 b0 60 60 s9 87

INTERNAL PRESSURE OF MOTOR, pala 2200 2500 8000 2800 2500 2500

CARBON MONOXIDEb, ppm 292 867 840 825 282 109

CARBON DIOXIDEbOc. ppm 22fl 2500 MW110 2500 1245 1505

NITRIC OXIDEb, ppm 4.2 3,0 8.6 3.5 2.2 43.0

NITROGEN DIOXIDEb, ppm ND ND NO ND ND ND

HYDROGEN CYANIDEb, ppm ND ND ND ND ND ND

AMMONIA, ppm ND 0.2 ND ND ND NO

TOTAL SUSPENDED PARTICULATE 71 63 71 70 67 NRMATTER, mg/m

3

LEAD mg/m 3 18 35 73 40 6.9 41.8

COPPER mg/mr3 2.0 838 9 1 4.4 4.0 4,8

ALUMINUM (a AL4O0) mg/rm ND ND ND ND ND ND

CHROMIUM mg/In 3 ND ND ND ND ND ND

ZIRCONIUM OXIDE mg/mr ND ND ND ND ND NO

a Nominal exft diameter was 1,0 Inches, However, th:s was in estUmate only, Actual diameter* could have varied between 0,75 and1.25 Inches.b Maximum observed conoentratbons,© Detemined In Runs 1.4 using Dreager Tubes, Runs 5 and 5 using NDIR analyer,d Special nozide used which Increased bum time. See teo Dais may not be represetnative.NR: Not RePiydedNO: Not Detected

20 0

TABLE 3

SUMMARY OF CHARACTERIZATION DATACOMPOSITION H

MAJOR CONSTITUENTS

RUN NUMBER 1 2 3 4

DATE 6,22-68 6.2288 6,22588 6,23-88

QUANTITY OF PROPELLANT, g 25 25 24 24

EXIT DIAMETER, Inches a I I 1

THROAT DIAMETER, Inche 0.261 0.261 0.261 0.261

ASOF CHAMBER TEMPERATURE, IF 70 70 70 72

ASOF RELATIVE HUMIDITY, % NR 68 57 83

INTERNAL PRESSURE OF MOTOR, plel 5000 5000 5000 50DO

CARBON MONOXIDEb, ppm 290 a 300 298

CARSON DIOXIDEb, ppm 250 0 270 290

NITRIC OXIDEb, pPm 4,5 a 1,7 5.0

NITROGEN DIOXIDEb, ppm ND 0 ND ND

HYDROGEN CYANIDEb, ppi ND a ND ND

HYDROGEN (HLORIDE, ppm -1 -I1

AMMONIAb, ppm ND NOD NO

TOTAL SUSPENDED PARTICULATE 87 0 73 176MATTER, mg/ma

LEAD mg/m3 0,771 0 06518 0,486

COPPER mg/mrn 0,726 a 0.807 0,500

ALUMINUM (as AL20,, mg/ma ND ND NDO

CHROMIUM mg/m3 ND 0 ND ND

ZIRCONIUM OXIDE mg/m3 NO o NO ND

MOLYBDENUM, mg/m 3 1,41 c 0.309 0.088

MAGNESIUM, mg/m3 0.261 a 0.224 0.250

TIN, mg/mo 0,348 0 0,397 0,177

a Nominal exit diameter was 1.0 inches. However, this was an estimate only. Actual diameters could have varied between 0,75and 1.25 Inches,

b Maximum observed concentrations,C Sample Acquisition fallure,NR: Not RecordedND: Not Detected

21

I II I

Table 4S

SUMMARY OF CHARACTERIZATION DATAOOMPO•• lO L

MAOf CONSTITUENTS

RUN NUMBER 1 2 a 4

Date 1.18-9 118.89 1.19.89 1.19.69

Quantlty of Propellarnt g 24 24 24 24

Exit Diameter, Inches a 1,0 1,0 1.0 1.0

Throat Diameter, Inches 0.28 0,28 0.28 0,28

ASOF Chanber Temperature, 'F 68 70 71 70

ASOF Flelati• Humidity, % NR 68 49 48

Internal Pressure of Motor, psla 2500 2500 2800 2500

Carbon Monoxideb, ppm 29B 337 371 371

Carbon Dioxideb, ppm 164 137 164 18O

Nitric Oxideb, ppm 1.5 0.5 0,5 0.5

Nitrogen Dioxideb, ppm ND ND ND ND

Hydrogen Oyamideb, ppm ND ND ND ND

Ammoniab, Ipm ND ND ND NO

Hydrogen Chloride, ppm 112 112 108 122

Total Suspended Particulate Matner, 50 33 38 51

Lead mg/mn3 2.73 2.71 1.2 1.50

Copper mg/o 3 5.74 4.43 3189 3.60

Aluminum (as A2O3) mg/m3 4.33 3862 3,35 3,14

Chromium mrn/m3 0,64 0,82 0.52 0.46

Zirconium Oxide m0g/n ND ND NO NO

Cadmilum, mg/m 3 0.15 0.13 0.12 0,11

C Nominal exit diameter was 1,0 Inohe. However, this was an estimate only, Aotual diameters could have variedbetween 0,75 wad 1.25 Inches.Maximum observed ooncentrations,

NR: Not PecordedND: Not Detected

22

Table 5

SUMMARY OF CHARACTERIZEATION DATACOMPOSITION a

MAJOR OONSTITUENT"

RUN NUMBER 1 2 3

Dats 12.140 12.549 12.549

Quantity of Propellent, g 65 64 60

Exit Diameter, Inches' 1,125 1125 1.125

Throat Diameter. Inches 0,18 0.190 0.197

ASCF Chamber Temperature, IF 66 63 64

ASCF Relative Humidity, % 34 46 40

Internal Pressure of Motor, pals 1580 1480 1100

Carbon Monoxideb, ppm 84 84 93

Curbon Dioxideb, ppm 1350 1324 1194

Nitric Oxideb, ppm 2 1 1

Nitrogen Dioxldob, ppm ND ND ND

Hydrogen Cyanldebl ppm ND ND ND

Ammonlab, ppm ND ND NO

Total Suspended Particulate Matter, 31 28 29mg/m

3

Lead mg/m3 18,6 1, 14,1

Copper mg/m3 0.002 000 0.01

Aluminum (as ALs20 mg/m.3 NO ND ND

Chromium mg/m3 0,0 0.02 0.02

Ziroonlum Oxide mg/m3 <0,1 <0,1 0.06

Iron, mg/rn3 0,33 0,06 0.06

"Nominal exit diameter was 1,0 Inches, However, this was an estimate only, Actual diameters couldhave varied between 0,75 and 1.25 Inches,

b Maximum observed concentrations,

NR: Not RecordedNr: Not Detected

23

TABLE 6

MEAN CONCENTRATIONSI ACHIEVED IN ASCF CUAMBER

Crtuant Pr tioilt Fom~toe sw xvt motor aize0

D__ _ _ b(75 9) N 50g) L (22 9) 0 (63 V) X (25 9

CO. ppm 330 295 344 85 195

CO-J4 ppm 1375 270 154 1250 561

NH2 p IMDL BMDL IMDL IMDL SMDL

NWO, ppm 3.5 4 0.75 1.3 5.0

NO,, ppm IMOL IMDL BMOL SMOL UMDL

HCN, ppm IMOL BMDL SMOL BMDL UMOL

HCL, pp IMOL '41 114 IMDL IMOL

Part Calls 70 100 43 30 45

Pb. mg,'m3 40 0.6 2 16 0.18

Cu, mG/rn3 4 0.7 4 0.01 0.45

AL,21, mgMr3 IMOL IMDL 3.5 IMOL IMDL

-Cr, mg/tr3 UML IMOL 10.5 10.01 1.3

Cd. mgm IMOL OMOL 0.13 IMOL OMDL

Sm. Mg/rn3 IMDL 0.3 IMDL IMOL SMDL

BMDL: Below method detection limit.

24

it was decided that determination of water vapor would be omitted from themeasurements.

A determination of the carbon balance for the chamber indicates that the analyticalmeasurements account for approximately 60% of the carbon in the formulation. Forexample, using the data in Table A-i for Composition D, there are ca, 2.06 moles ofcarbon in the motor. Data from Run 5 of the "D"test indicates ca, 1.2 moles of C tied upas the oxides of carbon (CO and C02). The analysis of the vapor and particle phaseorganic constituents (see below) indicates that only a very tiny amount of C is tied up Inthe trace species. And even if all the non-metal material collected as particulates was purecarbon, such would only add ca. 26 mg/m3 of carbon, or about 0.043 moles. Thus, it wouldappear that a significant fraction of the carbon present in the motor itself (ca. 33%) ispresent in some form which is not amenable to conventional analyses. Withoutconfirmatory data, the composition of such material would be highly speculative.

All of the formulations, despite the relatively small quantities of propellant fired In thechamber (ca. 1/7 to 1/20 of a typical size launch motor) produced substantialconcentrations of carbon monoxide, ranging from a low of about 300 ppm/100 g ofpropellant for Composition Q, to a high of nearly 1400 ppm/t00 g for Composition L.The amounts of carbon dioxide produced varied considerably, from more than a factor of10 greater than the CO produced, to only about half the amount of CO produced. Onlyvery small quantities of nitric oxide were produced, and no measurable amounts ofnitrogen dioxide were produced, The latter is not surprising, since the production of NO2is dependent on the square of the NO concentration 22, If the concentration of NO is low,significant amounts of the dioxide will not be produced in the first 10 minutes followingthe firing of the motor (the duration of time for which the ASCF was sampled for theoxides of nitrogen), Essentially, no aminonia or hydrogen cyanide was found at levelsgreater than 1 ppm. In the two formulations which contained perchlorates, measurablelevels of hydrogen chloride were found. However, the observed levels were notproportionate to amount of perchlorate present. For example, while Composition L hadabout 8x more perchlorate in the formulation than Composition H, the levels observed inthe chamber were about 100x larger. There were a number of metals found in theairborne particles resulting from motor firings, Copper, aluminum (as the oxide), lead, tin,chromium, and cadmium were all found in measureable amounts, Probably the lead andcadmium are of the greatest concern from a health risk standpoint. For bothCompositions D and 3 Q, lead was found to be present in the diluted exhaust at levelsgreater than 10 mg/m .

In Table 7 are listed the particle size distributions of the exhaust products for theformulations studied. The mass median aerodynamic diameters (MMAD) were all lessthan 2 um, indicating that the particles remaining airborne long enough to be collected bythe sampling method were capable of being inhaled. Although Composition D had ameasurably bimodal distribution, the higher of the two MMADs was still less than 5 'an.Particles from Composition 1. had a somewhat smaller MMAD than of the otherformulations, but the breadth of the distribution was larger.

25

I IIII

TABLE 7

Particle Size DistributionRocket Exhaust Particulate Matter

Mean Values

Mass Median Aerodynamic Diameter (MMAD) and Geometric Standard Deviation ({g)

Coriggitin MMAD (m) -QD& 1.46 1.86

H 1.44 1.77

L 0.807 2,14

Q 0.96 2A4

£ Composition D had a definite bimodal distribution:large particles had a MMAD of 3.6 microns, with a,0 1 .8;small particles had a MMAD of 0.47 microns, with o 1.7,

Trace Constituents

Trace organic vapor phase constituents present in the exhaust atmospheres weredetermined by collection of samples on multi-sorbent traps, followed by analysis by thermaldesorption GC/MS. Because of the sensitivity of the method, collection of sufficientsample was not difficult, However, the background levels of vapors in the chamber werevery high, and as a result, made it very difficult to discern quantities of vapors arising fromthe firing of the rocket motor. Despite the fact that the chamber was flushed with cleanair between most firings, background levels of collected constituents on chamber blankswere substantial (see Table 8). This suggests that there may be significant off-gassing ofvolatiles from materials adsorbed on the surfaces inside the chamber. Accuratequantitative determination of the constituents identified was exceedingly difficult, becauseit required determining the difference between two large values. Also, the largest peak

26

in many of the samples waw determined to be a mixture of hydrocarbons that were notresolved, even by high-resolution chromatography. These may be unburned, volatilizedwaxes used in the manufacture of the test motors. In Appendix B, in Tables B-I throughB-4, are listed the various trace organic vapor phase components identified and quantifiedin the exhaust. The data is summarized in Tables 9 - 12. In this case, mean quantitieswere reported only if the compound was observed in two or more of the traps analyzedfrom the firing of a specific composition and if the compound was present at a level 50%greater than the highest level reported for any blank collected during the series of firings.Several comments are in order. First, as stated above, it was very difficult to obtain a truly"clean" chamber atmosphere into which to fire the motors.

Table 8

CONCENTRATION OF SELECTED CONSTITUENTS IN CHAMBER BLANKS

Concentration Concentration ag&-3

C3-cyclopentane 52.4Methylene chloride 119 Ct 2-cyclohexasiloxane 8.2Methyl crotonate 2.1 C12-cyclnhexasiloxane 4.4C6-cyclotrisiloxane 239 C3-cyclopentane 7.4Cs-cyclotetrasiloxane 7 5 Diethylphthalate 19.1C3-cyclopentane 254 Pentadecane 2.1Terpinene 8.8 Nonadecane 2.6C1o-cyclopentasiloxane 129 Trimethylcyclobutanone 3.5Naphthalene 8.8

Originally, it was believed that the siloxane compounds may have resulted fromcontamination of the multi-sorbent traps with a soap bubble solution which was used inmeasuring the sample flow rates in some of the earlier studies. (This potential forcontamination has been confirmed by subsequent experiments in the laboratory).However, the siloxanes were also present in the blanks which were acquired in laterexperiments, in which only instrumental calibration of the flow rates were made. Thus, thesiloxanes may be off-gassed byproducts of the detergents used to clean the chamber priorto the motor firings, or they may be true products of the propellant combustion.Significant amounts of siloxane have been seen in the vapor phases of several of theexhausts from various motors. In general, there appeared to be a greater variety of traceorganics present in the vapor phase of the composition D and H exhausts, The fact thatComposition L is predominantly inorganic probably contributes to this observation.

Table 13 summarizes the maximum observed concentrations of non-siloxane compoundsfound in the ASCF atmospheres for those constituents with levels greater than 10 ug/m3(ca. 3 ppbv for benzene). For example, the average concentration for benzene was 17.6

27

Sg/m3 or 5.4 ppb. Overall, the concentrations of these species were several orders ofmagnitude below the levels at which they are regulated for workplace exposures, One mayconclude table 9

0

028

TABLE 9ESTIMATED CONCENTRATION OF TRACE VAPOR PHASE CONSTITUENTS

COMPOSlON D

CONSTITUEN'_r_ APPROXIMATE CONCENTRATIONS, cm3

Trichloroethane 0.4Benzene 13.5Trlchloroethylene 2,0Methyl crotonate 15.3Toluene 10.5C6.cyclotrlsiloxane 11C2-benzene 5,7Phenylacetylene 2,7Styrene 4.7C0"benzene 2.7C6-benzeite 3,9Decane 1.5Decane 0,9Terplnene 0.7C,.cyclotetraslloxane 15Terlpene 1,1Undecane 0,8Naphthalene 6.1C=.cyclopentane 1,3Dodecane 0,7C12.cyclohexasiloxane 17.8Hexadecane 1.1

a Estimated by determination of mean value for at least 2 of traps analyzed, which must be atleast 50% greater than the highest blank level observed, Levels have been corrected for blanks,

* 29

0TABLE 10

ESTIMATED CONCENTRATION OF TRACE VAPOR PHASE CONSTITUENTSCOMPOSmON H

CONS1TTUENT APPROXIMATE MEAN CONCENMTMON, valm

Trlchlorofluoromethane 9.8TrIchloroethane 0.4

Benzene 17.6Methylcrotonate 7.0

Toluene 2.2Phenylacetylene 2.4C.-benzene 0.7

Heptene 8.4Cyanobenzene 18.0C0-benzene 1,4

C3-cyolopentane 16.1C14.CyCloheptasiloxane 2.2

a Estimated by determination of mean value for at least 2 of traps analyzed, which must be at

least 60% greater than the highest blank level observed, Levels have been corrected for blanks,

30 O

TABLE 11ESTIMATED CONCENTRATION OF TRACE VAPOR PHASE CONSI'ITUENTS

COMPOSmON L

CONS.T'uJENT APPQ.OXMATE MEAN CONCENTRATIION, na/mr

Octamethyl-cyclotetrasiloxane 3.5Octamethy.cyclotetraslloxane 2.6

* Estimated by deteirnination of mean value for at least 2 of traps analyzed, which must be atleast 50% greater than the highest blank level observed, Levels have been corrected for blanks,

TABLE 12ESTIMATED CONCENTRATION OF TRACE VAPOR PHASE CONSTITUENTS

COMPOSmON Q

CO J •APPROXIMATE MEAN CONCENTRATIONA, L&m

trichlorofluromethane 0.6hexamethyl cyclotrlslloxane 0.2trlmethyl-cyclobutanane 23.5octamethyl-cyclotetraslloxane 0,3phthalate 8.5

* Estimated by determination of mean value for at least 2 of traps analyzed, which must be atleast 50% greater than the highest blank level observed, Levels have been corrected for blanks,

31

TABLE 13NON-SILOXANE VAPOR PHASE COMPOUNDS PRESENT IN

MOTOR EXHAUSTS AT CONCENTRATIONS GREATER THAN 10 ugtm3 In ASCFCHAMBER

Component Ma x i m u mConcentration. idm!

Benzene D,H 17.6Methylorotonkte H 15.3Toluene H 10.5Cyanobenzene H 18.0C•.-yolopentang H 16,1t2 methyl-cyclobutAnone Q 23.5

a Composition only listed if present at > 10Jg/m3 in that particular exhiaust atmosphere.

from this that the levels of trace organic vapor phase constituents are probably not ofconcern from a health risk standpoint under most conceivable use scenarios. Only byrepeated firings from an enclosed space could these materials reach toxic levels, Andbefore toxic levels of the organic vapor phase species was reached, CO levels wouldprobably be lethal.

Determination of the higher molecular weight particulate-phase constituents proveddifficult for the samples from the initial runs of Composition D (the first propellantstudied). Because of filter clogging immediately following the firing of the test motors, thenumber of particles collected was very small, For example, the largest amount of samplecollected on any of the initial runs was 40 mg. This was dispersed over a 4"-diameterTeflon-coated glass fiber filter, Initial GC analysis of the extracts indicated very low levelsof hydrocarbons. Next, the extracts were subjected to GC/MS analysis with selected ionmonitoring (SIM), SIM has the advantage of identifying species from selectedcharacteristic Ions, as opposed to using the entire ionic fragmentation pattern. Due to thesmall amounts of material collected on the filters, quantities detected in the particulatefilter extracts were considerably below our normal detection limits for the targetconstituents, For that reason, in the precedlV studies, the particulate collection systemwas modified to be a two-stage filter. This approach proved to be much more successfulat collecting greater amounts of particles, In Table 14 are listed the polynuclear aromatichydrocarbons (PAH's) determined in the exhaust particles collected from the firings ofCompositions D, H, L, and Q. In addition, a comparison is also made between theselevels and those determined for outside air at a military base, A few comments are inorder. First, only data for particles collected in the coarse filters are reported, The finefilters collected very few particles (1 - 5 mag), and thus many of the levels detcrmined are

32

at or neat the instrumental limits of detection. Nitro-PAHs were determined only forComposition D and H exhausts, The levels

33

10 '1 1*4 A~

ro I-

0 0i cc0 0 f

p ---- 34

determined in these earlier studies were so low that a repeat of the complex analyses didnot seem warranted. Despite the very low levels of PAH found in the particulates, theresults are fairly consistent from sample to sample. The concentrations of a few selectedPAHs in the particles of the Q exhaust were somewhat higher, but not by more than anorder of magnitude. The only nitro-PAH which was identified consistently in the exhaustsof the motors was 2-nitrofluorene, in the exhaust of Composition H. Its concentrationranged from ca. 30 - 60 ng/g. Most of the other PAHs identified and quantified in theexhausts were present at levels less than 1 ug/g. The outdoor air particulate sample withwhich a comparison is made was acquired outside a large motor pool building at FortCarson, Colorado, in the mid-1980's as background data for another project supported bythe USABRDL23. A major contributor to the particulates in this sample was expected tobe diesel- and gasoline-powered motor vehicle exhaust, The comparison indicates that,with the exception of 2-nitrofluorene, the PAH content of the rocket exhaust particulateis substantially less than (usually by a factor of 10 or so) that of outdoor air particulatematter found in a semi-urban setting at a military base, Also, the BaP content of theexhaust particulates Is about half that of cigarette smoke particulate matter24, Because ofthe relatively low concentrations of the PAH in the particle phase, the airborneconcentrations of the PAHs are very low. For example, at the maximum particleconcentration of '' mg/m 3 in the ASCF chamber (as a surrogate for human exposureconditions), the highest observed airborne benzo(a)pyrene concentrations would beapproximately 0.09 /.g/m3, and that of benzo(g,h,i)perylene would be 0,36/.g/m3 . At theselevels, the airborne PAHs and nitro-PAHs in the rocket exhaust probably do not representan additional health hazard above that of normal urban air particulates for the troopsusing such weapon systems.

SUMMARY AND RECOMMENDATIONS - PART I

The exhaust products from the firing of 2x2 rocket motors in a 20 m3 test chamber havebeen characterized. The data indicated that of all of the toxic and/or carcinogenic speciespresent, most were present at very low levels. Of the major toxic constituents, carbonmonoxide was the most universally present. Interestingly, the formulation with the greatestfraction of inorganic material (Composition L) yielded the highest concentration of COin the ASCF chamber per 100 g of propellant. Nitric oxide was present in all of theexhausts, but typically at levels less than 5 ppm in the 20 m3 chamber. No ammonia orhydrogen cyanide was observed at levels greater than I ppm. Levels of HCI were observedin the Composition L exhaust which were very high ( > 100ppm), and it seems likely thatfiring of this propellant in an enclosed space would produce very high concentrations ofthis toxic species. However, no data was obtained as to whether the HCI was present inthe particle or the vapor phase.

Particles were present at substantial levels in all of the exhaust atmospheres ( 030 mg/m 3).Particle size distributions indicated that for those particles which could be collected underthe sampling conditions employed, virtually all of the material was within an inhalable sizerange ( < 10 Arm mass median diameter). A large fraction of the airborne particles werecomprised of metallic species. Copper and lead (especially the latter) were present in theASCF atmospheres of many of the motor types at levels above those regulated by OSHA.

35

However, the levels of PAHs and nitro-PAHs in the particulates were very low.Comparison with airborne particulate matter collected at a military installation indicatedthat the PAH content of the particles was about 1/10 that of outdoor air particles.

Quantitative determination of the organic vapor phase constituents was very difficult dueto both the very low levels at which they were present and the presence of large amountsof other species In the background samples. Thie latter included a large number ofcyclosiloxanes, probably from the off-gassing of the chamber walls following cleaning. Onlya few exhaust components were found at levels greater than a few ppb. These includedbenzene, toluene, methylcrotonate, and cyanobenzene. These were typically present atlevels less than 10 ppb in the chamber.

From the standpoint of follow-on studies, recommendations depend on the goal of suchefforts. If the goal is to refine the comparison between the observed chemistry and thepredicted compositions, then the determination of methane (CH4) and molecular hydrogen(H2) would be very desirable. Such is a very difficult task, and would likely require adedicated real time mass spectrometer to make such measurements. However, thedetermination of such constituents would not significantly further the understanding ofpotential health risks of the exhaust products, since neither are toxic species.

Since these experimental studies were performed, there have been two developments Inthe field of analytical chemistry which, if' applied to these studies, could significantlyimprove the quality of the data generated, especially with regard to the determination ofvolatile organics. First, a number of carbon based adsorbents are now commerciallyavailable which have many fewer artifacts than the Tenax used in these studies. Were thesorbent traps used In these studies replaced with the new systems, it is likely that thenumber of artifacts present in the samples would be significantly reduced, minimizing thecomplexity of the interpretation of the data, Also, the recent development of directsampling ion trap mass spectrometry (DSITMS) for the determination of airborne vaporphase constituents is significant. DSITMS could '.. used to provide det%.rmination of anumber of volatile species of toxicologic interest in real time, much like an NDIR analyzerprovides real time measurement of CO or CO2. Transportable DSITMS systems are nowunder development at ORNL for air toxics monitoring at environmental remediation sites,and such technology could be useful for other scenarios.

Finally, the most important recommendation for future work is the determination of theexhaust product composition under actual field conditions, firing full scale motors. Thereare two important reasons for this. iirst, the data in this study indicates that changes inthe physical properties such as burn time can have a radical effect on exhaust composition.This suggests that it will be difficult to obtain highly realistic data unless true fieldmeasurements can be made. Secondly, firing of the test motors in an enclosed chambercauses significant run-to-run background contamination problems. Perhaps the firing ofmotors in single use, disposable structures, such as large nylon tents, would eliminate muchof the contamination problem.

36

S PART 2 - MODELING FOR HEALTH HAZARD PREDICTION

INTRODUCTION

Over the past 30 years, several digital computer programs have been developed at theNational Aeronautics and Space Administration's Lewis Research Center to carry out theconsiderable numerical calculations involved in the determination of the equilibriumcomposition of complex chemical mixtures at high temperatures•' 26,27. Updates to theseprograms have incorporated improved computational methods and adaptations toimprovements in computer speeds and capacities, In accordance with a suggestion fromproject management, we have used the 1986 version2s of the program described inReference 27 to obtain estimates of the composition of the exhaust gases from fourdifferent solid propellants. This was referred to as the NASA-Lewis model, version CET-86. The program obtains estimates of the equilibrium composition of a mixture of severalcomponents by minimizing either the Gibbs function or the Helmholtz function, Iftemperature and volume are constant, the Helmholtz function of a system decreases duringan irreversible process, becoming a minimum at equilibrium; if temperature and pressureare constant, the same is true of the Gibbs function2g, All gases are assumed to be ideal,even If small amounts of condensed species are present. Calculations can be done for anyone of six combinations of assigned state parameters (e,g.,temperature, pressure, density,entropy, and enthalpy); additionally, theoretical rocket performance data can be obtained,The assumptions involved in the calculation of rocket performance parameters are listedin Ref. 3. Briefly, they are: (1) validity of the one-dimensional form of the continuity,energy, and morentum equations; (2) zero velocity (no gas movement) in the combustionchamber; (3) complete combustion (in the sense that all reactants are converted toproducts); (4) adiabatic combustion; (5) isentropic (adiabatic and reversible) expansion;(6) homogeneous mixing; (7) ideal gas law; and (8) zero temperature and pressure lagsbetween condensed and gaseous species. An extensive discussion of these assumptions andtheir validity can be found in Reference 30.

The program first determines combustion properties in the rocket motor chamber andthen determines exhaust composition and properties at various stations in the nozzle,Since our propellants were fired in motors having a range of exit diameters, we used thefeature of the program that allows estimation of exit compositions for a set of several exitto throat area ratios. (In this case, the throat of the motor is considered to be the chokepoint, or opening of the smallest diameter, The exit is the exit of the motor nozzle, Usingthese definitions, the ratio of the exit:throat areas, Ae/At, must always be larger than 1.0.)In Table 15 are listed the ranges of exit/throat area ratios possible for each motor. Ineach of the predictions, we used the design pressure as the combustion chamber pressure.The throat pressure is defined to be the pressure at which the flow velocity is equal to thevelocity of sound.

The iterative procedures used by the program are discussed in detail in Reference 27.Briefly, combustion temperature and equilibrium compositions are determined for an

0 37

90

0. ,.. q-

I

3a8

assigned chamber pressure and the reactant enthalpy. From the combustion compositionsand temperature, the combustion entropy can be determined, Assuming isentropicexpansion, the program then obtains a first estimate for the ratio of chamber pressure tothroat pressure; from the throat pressure and the entropy, the actual gas velocity, thespeed of sound, and the Mach number can be calculated; if the Mach number is notsufficiently close to unity, the pressure ratio is corrected and a further calculation of Machnumber is done. Exit conditions for assigned exit-to-throat area ratios are also obtainedfrom an initial estimate of the ratio of the chamber pressure to the exit pressure, followedby iterative correction. The converged value of pressure ratio for each area ratio is usedas the initial estimate for the next area ratio.

We obtained the program, test case input, and output from the NASA Lewis ResearchCenter2s. We were able to compile the program on our VAX 6000.420 computer andwere able to reproduce the test case output with no problems, In our series of calculationsthe program has performed in a very reliable manner; we have had no difficulties with anyof the iterative procedures failing to converge.

RESULTS AND DISCUSSION

In Tables 16 - 19 are listed the predicted mole fractions of various exhaust componentsover the range of potential ratios of exit areas to throat areas. (The full computerprintouts for selected runs for each composition are included in Appendix C.) Note thatthere have been two independent checks of these computations 31 . First, CET86computations of mole fractions of Composition H were checked against the "Blake" codeand found to be in excellent agreement. (See discussion regarding Table 23, below),Secondly, the calculations were verified by running MUCET, a modified version of CET86prepared by Ell Freedman & Associates for use with microcomputers. Results wereidentical to those reported here.

The model has a cut-off feature. Essentially, it can predict the levels of over 100compounds, but will only report out those mole fractions which are larger than a user-specified value, For this work, a mole fraction of 5x10 7 was employed. The rationale forusing this value was as follows. If it is assumed that there are about 2 moles of exhaustproducts in the ASCF chamber following a firing, a mole fraction of 5xlO07 would beequivalent to lx l06moles of the particular product in the chamber. This assumption wasin fact supported by the chemical characterization data (see above). For a compound witha nominal molecular weight of 100 g/mole, this translates to a concentration of 5 1g/m 3 ,or 1.5 ppbv, in the 20 m3 ASCF chamber, Few airborne compounds are considered to bea significant health risk at such low concentrations, In addition, unless a very large sampleis acquired, it is usually difficult to confidently quantify such species at these low levels.

Using this criterion, with the exception of the metals in the exhaust products, the onlycompounds which were predicted to be present in the exhaust were carbon monoxide,carbon dioxide, hydrogen, water vapor, ammonia, and methane, In none of the cases didthe model predict significant quantities of nitric oxide, despite the fact that NO wasobserved at levels near to or greater than 1 ppm on each burn.

39

Table 16SPredicted Mole Fractions as a Function of Exit/Throat Area Ratios

CompoetOri DChamber pressure in OW0 p~al

AN 111300 118600 1 212500- 3,-j 1300 5.1700 6.2500

Exht T*-K 2258.4 1894,11 1788.5 IC 1.2618 1419.6 11355.0

______ ________ _______Mole fractions_________

Co .37059 .36871 .35390 .34478 .32876 .32241

coo .14561 .15759 .16241 .17154 .18756 .19391

H2 - .11245 .12448 .12931 .13844 .15445 .16080

H30 .23930 .22754 .22273 .21362 .19760 .19126

Cu(Total) 2.3949x1 04 2,40581004 2,4069A 1 O 2,406300 o 2,4063010*3 2,4062x I 0'

Pb(Totao 2,2823x1 0" 2.3222x1 0" 2,3276x100 2,332510 0 3 2,3352x1 0'* 2.336300*1

NIHO I.1 I09x 0., 8.7647x104 8,4223xi04 8.20800104 8.6068x104 8,8299x104O

00/00l 2,545 2Z276 2.179 2.010 1.783 1,663

NH/CO, 7,029x 10"' 5,562100" 85652100- 4,785,0lO' 4,589x 10'5 4,554x10

- Chamnber presure a 3= 0eAA .. 111300 1.8600 2.2500 3,1300 5.1700 6.2500Exit T,eK J2256.8 1893.7 11788.1 1628.4 11420,8 1355.7

_______ ______ Mole fraction.

00 .37001 .35869 .35m8 .34475 .32888 .32248

C02 14580 .15761 .1Il6243 .17156 .18744 .19384

H2 .11245 .12450 .12933 .13846 .15433 .16073

H230 .23933 .22752 .2271 .21359 .19772 .19133

Cu(TotaD 2,39608x 03 2,4059x1 03 2.4062x1 03 2,4634x1 Q3 2,4062x100 24063x100"

Pb(Total) 2,2819Ox 104 2.3219x1 0*3 2,327401 0'* 2,3322x 103 2.3355001 2.33650x1 -

NI-1 1,3315l00" 1,05119100 11.0110x00" 9,8554x104 1.0279x00" 1,O565xlO'

Co/co3 2.545 2.276 2.179 2.010 1.755 1,664

NH3/C03 9.1145100'a 6,67010-5 6.224X 10.8 5,745010'a 5,464x10'0 .5 0

A./A,: Ratio of the exit area to throat area

400

Table 17Predicted Mole Fractions as a Function of Exit/Throat Area Ratios

Composition H

Chamber pressure - 5000 psla

_/At 8.3000 10,000 16.000 23.000

Exit T,°K 1575,0 1607.1 1372,2 1251,4

Mole fractions

CO .25795 .25360 .24311 .23079

0O0 .25776 ,26229 .27332 .28608

Hg 8,5609x10 9.0087x10', .10095 .11357

H2O .24704 .24278 ,23242 .22018

HOI 4,5892x1 04 3.4824x104 1,8022x1 04 8.1443xl 0'B

KOI 1,3356x1 0' 1.2799x1 0' 1.0928x1 0' 7.7913x10'3

KCI(l)' 0.00000 0.0000 0 0.00000 1.5516100,

NH , 2,5247x1 04 2,5729x1 0'4 2.7684x 0'1 3.0523x0 0"

co/Co 1.0007 .9669 ,8895 .8067

H1I/0O0 1,7804x10"= 1.3277x 0" 6.5937xl 0"4 2,8469xl 0"

NHwCO2 9.7947x0 04 9,8094x1 04 1.01 29x1 0"' 1.0669xl 0"

A./Aj: Ratio of the exit area to throat area"1: Liquid

41

Table 18

Predicted Mole Fractions as a Function of Exltfrhroat Area Ratios

Composition L

Chamber pressure , 2500 psla

A__ _ 7,2000 10,000 15,000 20.000

Exit T, OK 1281.3 1175,4 1059,3 986.5

Mole frac.tlons

CO .14681 .13538 .11945 .10732

COo .11988 .13129 .14697 .16895

HCI .20072 .20084 .20139 .20167

H0 .25903 .24758 .23169 .21983

AI1O3 4.5708x1 0'" 4.5704X10'4 4,5672X10'" 4,6669x0 0"

BaC01(Total) 4,671x10'4 4,6849x10"4 4,6850x 104 4,6849x 0"

CrO(a) 8,1900x10' 8,1892xt0 4 8,1835x1 0*4 8,1831 X0l4

Ou(a) 0.0000 0 1,3642xl 0*' 8.3239x 04 1,1224X 10"a

NH•, 9,6149x1O" 1.0736xi 0" 11,2947x10'" 1.61 82x 10"

C0/00 1.228 1,031 0,813 0.675

HCI/CO, 1.674 1.530 1,370 1,269

NHaICO1 8,020x1,0" 8.177x1 0' 8.809x10'8 9,55661x10'

A./A,: Ratio of the exit area to throat area"k: Solid

"42

Table 19Predicted Mole Fractions as a Function of Exit/Throat Area Ratios

COMPOSITIN 0

CHAMBER PRESSURE - 1480 psia

__ __ __ __ __ __ 32.600 05.100 35.800

-Exit T, IK 918. 904.4 900.7

_______Mole Fractions_____

00 2.1 030xi 0. 2,0683x1 0" 2,0590x 10"

cot 1,8391x10' 1,8732x101 l.8823x1 0"

HI0 1 .0248x1 0" 9,9604x1 0 9.67J5x 10"

NH51 1.51 08X10' 1.5666axl0 1 ,581Ox1 0

Zr0, (Tot~al) 2.3203X1 0" 2,321 6x 1 V ~ 2.3220x1 0"

P b i.0228x 0,3 1.0234xI 0' 1.0236x 0'3

C H4 7.207,30 0'4 I 000SX10& 1 .088gx 1 0

BI 1.00656XI0" 1 3159X105 1.382Wx10

CO/e"02 1A143 1.102 1,094

NHd.C0, 8.21 6x10" 8.364i00'a2100

Aý/A,: Rath., of the exit area to throat aroa

0 /43

For many of the input parameters, the model was not particularly sensitive to substantialchanges. For example, for Composition H, a nearly 3-fold change in the exit/throat arearatios decreased the predicted mole fraction of CO by less than 12%. The ratio of majorcomponents was not signlficRntly altered. For Composition D, a 5-fold change in theA./At reduced the CO/CO 2 ratio by 35%. The ratios of minor to major components weretypically affected to a greater degree, In many cases, mistakes made in the original entryof data into the model were difficult to identify, since the mistaken or modified entryresulted In such a small change in the data output. For example, considerable effort wasplace into obtaining or calculating the best heats of formation for compounds present Inthe formulations. However, an exact value may not be particularly critical to the modelingprojections. For example, in Table 20 are compared the mole fractions predicted by themodel for a :5% change in the heat of formation of ammonium perchlorate, whichcomprises nearly 75% of the starting formulation. The results of the manipulation showonly minor changes in the predicted mole fractions. For example, the predicted molefraction of HCI changed only In the fourth decimal place.

From the standpoint of predicting the composition of the exhaust products in the chamber,the model was not particularly effective. As stated previously, in no case did the modelpredict NO to be present at levels above 10 ppb, even though NO levels wereexperimentally observed near I ppm. In Table 21 are compared the ranges of observedand predicted ratios of carbon monoxide to carbon dioxide in the ASCF chamber. ForComposition H, the predicted values were very close to those observed, For CompositionL, the model was accurate to within a factor of 2 - 3. For the other two formulationstested, there was substantial disparity between observed and predicted values, In both ofthese casee, the model predicted a much higher fraction of CO to be present than thatwhich was observed. If the model had been used to make a health risk projection, the riskfrom CO exposure would have been considerably overestimated,

The comparison of observed and predicted absolute concentration levels in the ASCFchamber is a much more complex task, Briefly, the moles of the elements present in theformulation were computed. Since we did not determine water vapor or hydrogen gas inthe chemical characterlzation studies, it was assumed that all of the H present in thoformulation was converted to water vapor. (From a functional standpoint of predicting theconcentrations of other species, it makes no difference If the H present existed as watervapor or H, gas.) Next, the total number of moles measured In the chamber wascalculated, assuming 100% efficiency of conversion of H to water in the chamber. Finally,the mole fractions of the various species were multiplied by the total number of molespresent, and divided by the chamber volume, in order to estimate chanmber concentrationsof the target species, The results of these calculations are summarized In Table 22. Ingeneral, the model was very good at predicting the concentrations of metallic species. Inthe case of zirconium oxide for Composition Q, and copper foi Composition D, there wassubstantial over-estimation of the concentrations, This may be due to settling ofparticulates containing

44

S TABLE 20Effect of 5 8% Shift in Heat of Formation of Ammonium Porchlorate

Composition L

Predicted Mole Fractions

H -' -74109, cal/mole

A./A, 7.2 10.0 15.0 20,0

Predicted Temperature, 'K 1248.8 1146.3 1033,7 963.8

CO .14393 .13194 .11561 .10325

CO0 ...... .12259 .13431 .15041 .16255

OO/CO2 1,17 .98 .77 .64

HO .26526 .24320 .22699 .21523

Hl .19284 .20402 .21948 .23068

HCI .19924 .19992 .20044 ,20078

N, 7.833xl 0' 7,826xl 0" 7.822xt0" 7.823x10"

Cu(s) 1,583X10"* 2,442xl 0"3 3,070x0 0 3,331x10"

NH3 1 ,143x1 0'5 1,284x10"' 1.566x1 0" 1,836x1 0'"

H, = .67051. cal/mole,JA__ _ 7.2 10.0 15,0 20.0

Predicted Temperature, 'K 1300.9 1194.0 1075.7 1001.2

CO .14912 .13778 .12215 .11017

CO2 .11748 .12854 .14394 .15578

CO/CO3 1,27 1.07 .85 .71

HaO .26048 .24902 .23337 .22157

H2 .18794 .19841 .21330 .22469

HCI ,19898 .19975 .20032 .20059

N2 7,836x10' 7,827x1 0" 7,821 xl 0" 7,820x1N0"

Cu(s) 1,2.7.x0o 2,240x10 1 2,958x 10" 3,260x10"

NH3 8.885x10"0 9,774x10 4 1,169x10"' 1,367x10'a

.!A3 Ratlo of the exft area to I roat area

45

these species before they could be collected, For Compositions D and Q, the modelsubstantially over-predicts CO and underestimates the amount of CO2 produced, In thecases of the formulations which were expected to produce measurable amounts of HCI, themodel predicted more HCI than was measured in both cases- It could be that in this case,the acquisition of the sample could be suspect. First, some of the HCI or potassiumchloride could have been adsorbed on particulate matter which settled very rapidly in thechamber, In this case, the material would not reach the input to the continuous HCIanalyzer. In addition, some of the HCI may have been lost in the short lengths of Teflontubing leading from the chamber atmosphere to the analyzer,

TABLE 21

COMPARISON OF OBSERVED AND PREDICTEDCARBON MONOXIDE: CARBON DIOXIDE RATIOS

Observed Predicted

Propellant Composition Minimum Maximum &JInimunm Maximum

D 0,0924 0.2265 1.663 2.545

H 1.028 1.160 0.8067 1.0007

L 1,817 2,473 0.675 1,225

Q 0.0622 0,0779 1.094 1.143

In terms of the trace organic vapor and particle phase constituents, the model correctlypredicts that the concentrations of these species will be low, In fact, the observed levelsof such species as benzene and benzo(a)pyrene were much less than 100 ppbv, or 1 jUg/m 3,respectively. However, the number of toxic species which the model considers is limited,and it is certainly conceivable that a compound riot considered by the model could bepresent at sufficiently high levels to warrant some health risk consideration,

LIMITATIONS AND MODIFICATIONS

In addition to not considering all of the toxic species likely to be produced by the Ignitionof a predominantly organic matrix, the model does have several limitations, First, it Is anequilibrium based system, and does not take into account those synthesis pathways which

46

may be governed predominantly by kinetic processes. Secondly, it assumes ideal gasbehavior on the part of all of the gases produced. This assumption is not likely to beaccurate over the entire range of conditions existing inside the rocket motor. However,from a practical standpoint, this may not be a severe limitation, For example, themagnitude of non-ideal gas effects depends primarily on the density and the temperaturein the system. For the system in question, the largest densities occur in the chamber.Interestingly, the most dense gas (H), has a density of only 0,037 g/mL, which Is notsufficiently large to induce substantial deviations from the ideal gas law, To Illustrate thispoint, Freedman31 has used the "Blake" code to compute chamber concentrations (at340,23 atmospheres pressure and a temperature of 3167' K) assuming both ideal and realgas equations of state. This was performed for Composition H, whose exhaust productswere capable of reaching some of the higher temperatures in the study. The results arelisted in Table 23, It is clear that the differences between the real and the ideal gaseousequations of state are very small, And although there are differences between the NASA-Lewis results and those from the "Blake" code, the differences are negligible from apractical standpoint and are due to differences in the thermodynamic data basesthemselves.

Finally, and probably most importantly, the model assumes that all of the chemicalprocesses are frozen at the point at which the exhaust gases exit the motor, There is aconsiderable body of evidence to suggest that this Is not the case, For example, the modelpredicts that no significant production of NO will occur for any of the formulations tested.However, NO was In fact observed. We believe that its presence is due to the effect ofthe heated exhaust gases on the ambient air in the chamber, That is, the heat from themotor firing causes the formation of nitrogen monoxide. The production of NO isprobably proportional to the duration of the flame contact with the air, For example,during run No, 5 for Composition D, the shock wave from the firing of the motor causedsome damage to the chamber. A different nozzle was installed on the test motor used forburn #6, This lengthened the burn time, and reduced the pressure of the burn, Suchresulted in some substantial differences between burns #5 and #6 for the Composition Dmotors. The change in the NO concentration is considerable. Probably, the Increase istime that the flame is in contact with the air causes much more NO to be produced. Notealso the change in the CO concentration from Run No, 5 to Run No, 6,

Following consultations with Dr. Eli Freedman, we decided to test the hypothesis thatincluding a step in the computer calculations which would determine the influence ofmixing the predicted exhaust gases with ambient air would lead to a more accurateprediction of the observed gas concentrations in the chamber. The model was revised tomix the exhaust gases with the ambient air at fixed ratios and at varying pressures andtemperatures. As an example, the exit composition from propellant D (a formula whichhad initially yielded a relatively inaccurate prediction of the observed CO/CO2, ratio) wasselected as a "fuel" which could be mixed with air, Initial exit pressure and temperaturewere set at 39.5 atmospheres and 1837 *K, respectively. The "fuel" was mixed withambient air In the ratios given in Table 24 to yield equilibrium compositions at twoarbitrarily selected lower pressures. As indicated in Table 24, there was a substantialdecrease in the CO/CO2 ratio, The resulting ratio is much closer to that which was

0 47

Sobserved experimentally than the ratio predicted by the unmodified model, suggesting thatthere is considerable mixture with ambient air and conversion of carbon monoxide tocarbon dioxide between the vicinity of the motor exit and the analysis train, That themoodel does not cowsider the influence of mixing with ambient

0

ii 48

E

0

E-Ri *

0~ 0 C

8' --Chr

49

TABLE 23

Effect of Choice of Gaseous Equation of State on Computed Mole Fractions forComposition H'

BLAKE NASA-Lewis

NAME IDEAL REAL IDEAL

CO 0,2928486 0,2932262 0,29422

H. . 0.2679565 0.26,,77 0,27100

CC0 02183805 0,2180917 0.21722

N2 0.1346118 0,1,346414 0,13459

H2 4.927155 x 10' 4,886758 x 10" 4,8588 x 10',

HCi 8,636553 x 10'* 8,599959 x 10.*

KOH 7,788912 x 10" 7.757804 x I0'_

KOI 7,232547 x 10'3 7,278343 x 10"

NO 1.281355 x 10'3 1.270143 x 10"0

02 5,7927.95 x 104 5.639095 x 10-1

NH0 8.57131 x 10"6 8.776596 x 10"4 _ _,

CH=Q 2,823712 x 104 2.871074 x 10" _

HCN 2.529327 x 10"- 2,631338 x 10"4

C12 2,863636 x 10 .' 2.811794xl 01.

COCI2 2,512875 x 10"° 2.628192 x 1101°

K 1,164592 x 10"3 1.15023 x 103 8,4006 x 10*'

COCI 1.79761 x 10"6 1,84523 x 10"'

OH 6.396093 x 10'3 6,222507 x 10'3

KO 5,224935 x 10" 5.182151 x 10".

H 3.155921 x 10"3 3,057469 x 10"

0 2,448266 x 10,4 2,370879 x 10"4

N 1.259862 x 10a 1,24317 x 10"6

CHO 2.055275 x 10-5 2.080149 x 10 _

Cl 3,638269 x 104 3.574871 x 10"4

From Re erence No. 30

50

air on the products of propellant firing has been observed by other investigators32.Sneison, et al, reported that double base propellants fired in Argon atmospheres producedmole fractions of CO which were much closer to, those predicted by thermodynamicmodeling than when the same propellants were fired in ambient air.

Table 24

Influence of Exhaust Gas Mixing with Airon Carbon Monoxide/Carbon Dioxide Ratios

Composition D

Fuel/Air = 5*

Pressure, atm 39.5 5.0 1.0

Temperature, OK 1837 1300 1000

CO/CO2 1.44 1.08 0,74

Fu~i/Air - 3*

Pressure, rtm 39,5 5.0 1.0

Temperature, *K 1837 1300 1000

CO/CO2 1.16 0.88 0.61

Fuel/Air 1 _

Pressure, atm 39,5 5,0 1.0

Temperature, *K 1837 1300 1000

CO/CO2 0.31 0.25 0,17

* Considers exhaust gases from motor nozzle as "fuel,"

mS

RECOMMENDATIONS FOR FURTHER WORK

It would be interesting to compare these results with other computer models. Softwareis available with similar, but not identical methods of computation and data fitting33.

It may be possible to extend the NASA Lewis model to account for nonideal gas equationsof state for some of the major components, without involving major modifications to theprogram. However, any revision is not to be undertaken lightly; the program is some 5000lines of Fortran and represents a very large investment of time and effort. Thedevelopment of a new model would require a similar investment.

A thorough review of the thermal and transport property data base may seem to bedesirable, in order to incorporate any new information available since the 1986 revision,and to have some additional assurance that the data have been entered correctly.However, there have only been 8 changes to the data base, and none have practicalsignificance for this study3?. And since transport properties are not a significant factor inthis work, any changes should not have an effect on the conclusions.

It would be useful to model the chemical kinetics of these processes, using the softwaredescribed in Reference 34, It should be noted, however, that a considerable amount ofeffort would be required to elucidate the reactions occurring in these events and to makeestimates of the necessary rate constants. The Arrhenlus constants and the activationenergies for the hundreds of conversions processes are not available. In contrast, modelingthe flow processes may be useful, since it could lead to a better understanding of theamount of air entrained with the exhaust during combustion.

It might be useful to do some experimental firings of the motors into inert atmospheres,such as argon, in order to test the air mixing hypothesis, However, such in and of itselfwould not aid in the refinement of the model.

Finally, alternatives to the "air entrainment" explanation as the source of disagreementbetween experiment and computation should be explored. For example, calculationsdescribed in this report were carried out for two possible cases: either the chemicalreactions in the expanding flow from the combustion chamber maintain completeequilibrium from throat to the nozzle exit, or else the flow is completely frozen once itleaves the nozzle throat, But the intermediate case is also possible. That is, the flow mayfireeze somewhere between the throat and the exit, This could provide a possibleexplanation for the discrepancy between experiment and computation without requiringthe assumption of entrained air. To implement such an approach, an adiabatic expansioncalculation should be run. Initial estimates provided to the authors of this report suggestthat this approach is feasible31 . However, to take full advantage of such an approach,careful experimental determination of hydrogen and methane would have to be performed.Because of the complexities of such real time analyses, these measurements could not beperformed.

052

0 REFERENCES

1. FM 6-20 (with Cl), Fire Support in Combined Arms Operations, 30 Sep 1977.

2. FM 71-101, Infantry, Airborne, and Air Assault Division Operations, 26 Mar 1980.

3. AR 40-10, Health Hazard Assessment Program in Support of the Army MaterielAcquisition Decision Process, 15 Oct 1983.

4. AR 1000-1, Basic Policies for Systems Acquisition, 1 June 1983.

S. Characterization of Combustion Products from Military Propellants, lIT ResearchInstitute, USASMRDC Contract DAMDI7-80-C.0019.

6. Short-Term Intermittent Exposure to HCi (Draft Final Report), Enviro Control,Inc., USAMRDC Contract DAMD17-79-C-9125.

7. Hoke, S.H. and I.W. Carroll, Development and Evaluation of Atmospheric HCIMonitors, In Toxic Vapor Detection Technology (Propellants and Related Items)S&EPS Workshop, CPIA Publication 386, October 1983.

8. Letter, SGRD-UBG-M, 30 April 1984, subject: Medical Research IssuesAssociated with Stinger.

9, Letter, SGRD-UBG-M, 18 Oct 1984, subject: Development of a CoordinatedMethodology Investigation/Medical Research Program for Evaluation of Gun andRocket Combustion Products.

10, Letter, (2nd End), SGRD-UBG-M, 24 Feb 1984, subject: Stinger Exhaust GasMeasurement, TECOM Project No. 3-MI-000-MAN-031,

11, MFR, DASG-PSP, 28Jan 1982,subject: RCI Health Hazard Assessment, MultipleLaunch Rocket System,

12. Minutes of Meeting, Standardization of Test and Evaluation Procedures forChemical Hazards in New Material, 13-14 Octo 1982, DASG-PSP, dated 15 Oct1982.

13. Letter, SGRD-PLC, 15Jul 1981,subject: HELLFIRE Human Factors EngineeringAnalysis.

14. Letter, SGRD-OP, 4 April 1979, subject: US ROLAND, Health HazardAssessment, ASARC 11I.

15. Lett-, DRXHE-MI, 31 Oct 1980, subject: US ROLAND, Health HazardAssessment, ASARC lIllb.

53

16. Letter, SGRD-OP, 15 Feb 1979, subject: Health Hazard Assessment, USROLAND, ARARC III.

17. Letter, DRSTE-CM-A, 15 Oct 1980, subject: Safety Release (Limited) for RAMDemonstration of US ROLAND at Ft. Lewis, WA.

18. Letter, STEWS-TE-MF, 2 Aug 1982, subject: Exhaust Gas Measurements onSTINGER Firings,

19. Letter, STEWS-TE-MF, 21 June 1983, subject: STINGER Exhaust GasMeasurements Test Plan (TECOM Project 3-M-OCO-MAN-031, andendorsements (2) thereto.

20. Letter, STEWS-TE-RE, 14 Feb 1983, subject: 1ICI Gas From STINGER Firings.

21. Keith J. Laidler, hemjical Kinetics, McGraw-Hill Book Company, New York, 1965,(pp138)

22, R. A. Jenkins, C. V. Thompson, T. M. Gayle, C. Y. Ma, and B. A. Tomkins,Interim Report, "Characterization of Rocket Propellant Combustion Products -Description of Sampling and Analysis Methods for Rocket ExhaustCharacterization Studies," ORNL/TM-1 1643, June 7, 1990.

23. W. H. Griest, R. A, Jenkins, B, A, Tomkins, J, H. Moneyhun, R, H, ligner,T. M. Gayle, C. E, Higgins, and M. R. Guerin, Final Report, "Sampling andAnalysis of Diesei Engine Exhaust and the Motor Pool Workplace Atmosphere,"ORNL/TM-10689, March 1, 1988, DTIC No. AD-A198464

24. B. A. Tomkins, R, A. Jenkins, W. H, Griest, R, R, Reagan, and S, K. Holladay,"Liquid Chromatographic Determination of Benzo(a)pyrene in Total ParticulateMatter of Cigarette Smoke," J. Assoc. Off. Anal. Chem. (A(5), 935-940 (1985).

25. F.J. Zeleznik and S. Gordon, Calculation of Complex Chemical Equilibria, Ind,Eng. Chem, 60, 27-57(1960).

26. F.J. Zeleznlk and S. Gordon, A General IBM 704 or 7090 Computer Program forComputation of Chemical Equilibrium Compositions, Rocket Performance, andChapman-Jouguet Detonations, NASA TN D-1454, 1962,

27. S. Gordon and B.J. McBride, Computer Program for Calculation of ComplexChemical Equilibrium Compositions, Rocket Performance, Incident and ReflectedShocks, and Chapman-Jouguet Detonations, NASA SP-273, 1971; interim revision,1976.

28. B.J. McBride, personal communication.

054

29, M.W. Zemansky, Heat and Thermodynamij, Fifth Edition, McGraw-Hill BookCompany, New York, 1968. (pp. 279-281, 557-606).

30. D. Straub, Thermofluiddynamics of Optimized Rocket Propulsions, Extended LewisCode Fundamentals, Birkhaeuser Verlag, Basel, 1989.

31. Ell Freedman, Personal communication to Steve Hoke, USABRDL, February 12,1991

32. A, Snelson, P. Ase, W. Bock, and R. Butler; "Characterization of CombustionProducts of Military Propellants, FINAL REPORT, Volume II,1 AD-A167417,March, 1983

33. R.J. Kee, personal communication.

34. R.J. Kee, J.A. Miller, and T.H. Jefferson, CHEMKIN: A General-Purpose,Problem-Independent, Transportable Fortran Chemical Kinetics Code Package,Sandia National Laboratories Report SAND 80-8003, March 1980; A.E. Lutz, R.J.Kee, and J,A, Miller, SENKIN: A Fortran Program for Predicting HomogeneousGas Phase Chemical Kinetics with Sensitivity Analysis, Sandia NationalLaboratories Report SAND 87-8248, February 1988.

55

0

Appendix A

Selected Rocket Propellant Formulations 0

556

17

I-S

I Ii I

.57

Table A-2

COMPOSmON 'H9 FORMULATION

Abbreviation Constituent Formula Wt % 140 (kcal/mole)

KCIO4 Potassium perchlorate KCIO4 7.8.8,05 -103.43

NC Nitrocellulose CI2HjNBOKg 54.60 169.17

NG Nitroglycerine C3HON 304 35150 -88,6

EC Ethyl Centralite CIIHNgO 0,9-0,8 .25.1

C Carbon Black C 1.20 Ref.

The entry 'Ref,' In the heat of formulation column means that this is a reference element in theNASA-Lewis program.

58

Table A.3

COMPOSmON 'L" FORMULATION

Abbrviatlon Constituent Formula Wt. % AHr. (kcal/mole)

AP Ammonium Perchloat. NH4 0104 73.93 -70.58

PVC Polyvinyl Chloride A(0 H Cl) 11.67 841

DEHA DI (2-ethyl hexy!) adipate ... 2C Hg 04 11,67 -308,0

CUOCR Copper chronite .. CuC2r, 04 0.97 Rt,.

Al Aluminum Powder Al 0,99 Rot.

C Carbon Black C 0,05 Ref.

BACD Stabilizer Be-Cd 0,47 Ref.(Barium/Cadmlum)

SOSS Sodium dloctyl sulfo 0,1,0 H31 0 SNa 0.083auccInate

GMO Glycerol monooleate C,1 H12 04 0,083

PTD Pentaerythrltal dioleate 04, H7, 0 % 0,084

0 _* Heat of formation unavailable

59

Table A-4

PROPELLANT '0 FORMULA11ON

Conethtunt Formula Woightd

NO NitroglyoorIna 3 5N0 11.38 -8860

877N Butane triol trinhratu 041-1N309 11,36 -03.07

HMX Cyclotstromethylone 04H3IN608 W600 17,93tvtranItram~n*

PGA Polyglycol adipale C10H1605 4.63 -282.9

WOO0 Tri-funotlonai Isocyniate C1 H3N0 1.66 -23.55

MNA N-methyi-p-nftroaniline CHIN1,Op 0.76

4.NDPA 4.nltrodiphtnylamine Oj.H 11N2,0 0.40 15.4

pop Polyesproiaiotono polyl C60?2 0.34 -858.1

NO Nhtroceilulotee C,3 H1,N8O1 0.34 169.17

Lead Citrate Pb3(05- 1507)2*3H20 1.10

zrC Zirconium Carbide Zra 1.00 .48.5

C Carbon Black C 0.40 net.

TIDE Triphenyl bismuth 8l(061-1ý 0.04

The entry 'Mef,' In the hest of formulation columtn means that this Is a reference element In the NASA-Lewis program

Heals of formation unavailable

60

0

Appendix B

Trace Organic Vapor Phase Constituents ObservedIn Selected Rocket Exhaust Atmospheres

661

IL

a - o -

0

++°~ Ca,+ ++62

=1 - = = = - = =

63N

-J - i = - -- ===

~00LI0

Cd- w

64

J 1 I... .. L

64

IL

65 °

I•,--

i , ,

iI

111=il L iS - - - - - - . - - === -

a6

H.

I N

C. 0y

66

SIi i

I°

• ij !!.!~ d •II.' 7

Table B-3Trace Organic Vapor Phase Constituents

In ASCF Chamber

Compositon 0

Wt TW MTl. Y1M KAM(. MM ThTAA ~WAID- ~TWA& IWrM. MUAkwinbu mm BL"4 .1 a a 4 a U1

Btoo _______ 032 .420 Ile" 0.150 007 1,7117 25811

carboni dioxide 14 0.720 1.6"

trixhlo Obinfum~eltham 1101 01016 Cal?

OCIothmUyl-YI'cciNVrAIStAOt at.0 11490

heaMnetylcyI~cIottiallOaxaat 211 cOMI 0.00 Ilial o.07

hexame1vilc10,eIoaMw a.? 0.048

hexamiethyl-eyal~etbuJ~me 1316 0.044

tJirnethyliSIIM .mpd 140 05650

coa I~yott~ibmn 55 04214 0.050 0.501140

hyftcarbom 17A 0.057 0.402

anJIylacohoto 27. 2 , 175

heXaMethl-cIyclotdioxan'e 3306

decamethyl- go' 0.01 0.074 0.277

naphthalene 31.1 0.07i

trimethyI-cyclabutarmone 31.6 0058 61162 0.43 10.024 3.504

hemam.1hyI-eyuaiotrsIoxanie 34 0 0.030

Detarnothyl-cyclot~lasiluxmsi 3006 0.060 0,423

phthwaa 30.2 14.20 0064 7A220

h~xAmethylIacyoetrisiloxgn 42,3 4MGW

plhth&isl. 43.9 00611

68

Appendix C

Output from Selected Runs of Computer ModelNASA-Lewis CET-86

69

0

Table C-1

NASA - Lewis CET -86

OutputComposition D

7

70

cc,.. Ce,., tee,.ccc.. C..,, eec,.e.e.c I.,.. ,,cccccc,. cc... e.e.ccc... e.e.c cc..,eec., cC.,, eec..cc,,. cc...

ccc.. cc...cc... ccc.. cc...e.e.c ecc.. eec...cce. ccc..etc.. eec,, ,cc.cS..,. a..,. Ce...e.e.c ccc,. ecce.cc... cc,., cc...cc.,. ccc., etc..

cC.., cC... ccc..cc... Cc.,. e.e.cetc.. cc.,. cect.e.e.c cc,,. cc,..e�cee etc., e.e.ceec.. ccc.. e.e.ce.e.c e.e.c e.e.ce.g.. Ce.,, cc,,.

e.e.c cc,,, tOS*Se.e.c ccc,. eec,,ccc.. e.e.c ecceeeccee e.e.c etee.see.., ecte,eec.. ccc,. e.e.cecece e.e.c cee..eec.. 9.,,. cc..,eec,, ccc,,eec.. eec,. cc..,mc... ccc,. ccc..tee.. c.e.t eec..see., eec.. Ce,..tee.. e.e.cI I

* �eeeeec.11111 cete!

eec,, em.,. *eccS.,., ectec ecet.e.e.c tee.. ci....ee.c cc...

cc.,. etc..bee.. ci...

cc,., e.e.c see..iii�i �a..., tctte a a a eec..

cc,.. a a ececee.e.ccc,.. a *tcceccc.. e.e.c

ii::: ** IH.ii, IIH!eec.. e.e.c � cc...

ececeS.... etc., Ce..,eccec eec., .c.ece.e.c eec,. eel..e.e.c eccec eta cacceccc.. ccc., eta e.e.cccc.. a.,,. tat eec..Cc... cc.., ccc..cc... ccc.,cc.,. eec,. ,c.ccce... etc., a e tececeec.. ccc.c a a .e..eccc.. cc,.. a a ececteec., cc,,. at.... ccc..tee,. tee,, eec,,Ce... e.e.c becc... e.e.c a a eeeccetc.. cecte a a *cct.

* eece tttaat cCc.*cec.. ccc,. tac..

o tee,, a cc.,. a e.e.cU emeta a ecec. e, cc,..

cc.,. C tcce, a ccc..* tee.. teatta e.e.c aataaa eec..

.ce..m ecece ace.,ci ccc.. cc.., e.e.c

ccc.. a ceme. t a etc.c* eec., a ccc,, a a cc.,.- eec,, a cc.,, a a cc..,

ccc.. a ccc,. a a eccec4 ceeec cRUet. cecee atwa ceece

*.ece ccc.. cc...tee.. ccc,, ccc.,

C cecec a a ccc.. a a ,cc..* cc,.. a a ccc,. a a .,,ceU cccc t a ececc a a e.e.c

eecc. a S ccc., a a eec..- ecece a a ccc,, a a c.ccccc ccc,, eta. eccec SaRa teec,� eccec cc.,. ,ccc,

* 71

-w"am

o 0 WL

U L

8& 4 aa w~

O W UN. P 4 4 4

u a*4I444 a ai w

40 a L'.

00000400

:0 a 0 M.*le .1 d 4 .J .14 '4 P U

~nI I II 4

a no hi 4*u4 44I W4 L

444444O 4W .44Ua .a444444*4 444 11- 4

c34 4 4 2 w ..

L~~~UU WWOaU0A AaZ~~~~N '0 -K 091 AM L.L& . W& %4 .44 3

aUUmUUUUL 4444444434Gý

4444444 '.'.N 72

a411 U aaAE U U W AI A

!.!b3U ..

.4.LU A U U ft 0 "U#

UPA *A4UAUA .UIn "a

..* I*.44. u s m o m ! wwUw 00. t,

45 %A AA AAA AAAN .4E.A *

A 0 0440.

UsA-~A: -a

1% AM &a .4 0 &-

ONAUM.4AA U .s40I1

A WUA~h AN A A EAA.4Ann

USAA UMW

a5 5 55 ' . & . U a aMa. L

* 4.44 b.A U.433U A 49W ." O U A SA

AM AM 4AW -ll ....MP& U

WA~ ~ ~~~~ 1.U A IA -U 11 I I I II I I

EU WU UI U 349 434NU U- 04 --U - d o oMSI" 4.US U* 0$ AU UA. UU

SES UI. LASU U U a

73 U bUUU

UN MC N N U --Mo *U 4 * 4 -rnu N. a * U EU

N. MU U U U EU-� 4- @4 U- - inNIi II hi hi I LII

CU 4N C N N MM:: �.�N M P*iM

N,,44 44 '.4 .14 4 .44

I I Ut Mi I WI IU U U

ON 44 4 4 N 44

UM NA 4 4 44

44 CC C C 4 44NM 044 MN WIN N $iNN

Ii II ii I III

ON Np. N 4 4 44

NE MV. U N 4 CM

UN WON U U 4 US

us S. S UN p. uNN*4 '44 p.4 MC C Netii Mi ml i iii

Np 4?' A' - 4 MNEN 55 SI 4 N *4I�4 CV. - N S NP

MC 44 NC 4 4 *�4444 44 54 54 C CCC

II Ut Ut I WIt

MM Np. - 4 4 44NM V.4 U U C SC

MV. NV. 4 4 4 4444 44 SM NM N 044

44 CC M N S CE

4N MN N C * CV.

*4 44 4 MMMM

44S44 34 UMIANU INM

* NI *4 U W -

I. N N N4?'5V. CE SCOUt BUS

o - In 11 U -C C U C S U

h & A

74

14.4K Z U0 0 1 K .

1.44

a u m .&N is4. .4 pftw

- ue wa n m Ue I ! aNo M-u A

a I., PA0

UON ~ ~ ~ ~ ~ ~ ~ ~ ~ s as .N U . 4 %fVN4I C .K'NN N 4 4 S . .

a N U a aa , an .45 mq.VIN6 aK 44. a 4N 1.tP a O.UN.U.

IV I'l a s . 0. I .44514.N44N4NN

K ~ ~ V. 41 43E6141 WN1iV 0N.U o- 4 4NENN61U4Ua

N 4 a44 a oa-ladi

*W 4

4. K~uW 1 N.unNn NMNS UNN 4N..464U .61.4 ,46 WP E U K U4 isI a a I

- 4 UVE .. N 1K VI4 4N6N N6KN1 noa4K U~~~~~~~~ 45 .604IN - U UN U4.4 24.4VK1KMEUU K 414.E.4PI O tamJ .," 'K N l%14U6UU iE4.4~~~~~ a4 N*41U. lt 14.4 WaE N N M 4 N M

.4 '4.4475

*~~ ~ aa -..

0 a4 46I *

- ý L 4. a6 3U

* ~ ~ ~ I 43 0 4 a a .

a i a

N ~ ~ 6 a IS. NU I4 " Uao

a. a 0 U. 03 UN L

Oil I.is t u 6

* ~ ~ ~ ~ ~ .666 .66646 5

00436.6f ft p6I A5I.03 6W3I464

Li 34S44t 44U164676

uxz,&a

""N 9 0 L5.4 0 RWUIN04 ONN hit S

W4 000 0.W S.WN-.06 8 0 .iit aW

mosu SJ a SUa

P' 4 .. too -a

4 0I -AWwII N .4au momWW VS

- hi hMoo.in hiIt

U~4 10U W

-. ~~~~0 L.ad~o CIW W N

*A . 9... . is i.Uih aO N

S ~ ~~~~ LNa~oU I I Z S Uhi~NN~iS C .uxozooU&

a aS400 13 0 WN ai4nNUS II hinhi WUNESUN S UWSU&Uuz

el 10NN .0 .

O1 . h .. , S N u5 N u .. 4 I.IS dm . .i

a 4JN ai IN , u 1, N'M A4N a I a Ul 1

* SUUN NI%.4 ~ hiOL, NSi 4USU US i S hi17

U

1h U

Ut KESOW Sb

4

K ..'

bEE 0 �hi'.*� U hi'.-953U 44-OUIi'S 3�UNU.4 3 U U U 3 U Kb

"aU

.,La- *44�004 4�4�it I�4� la-

*hiWN'S�iI.S�i -U'SflUUEIiUL U

Ua

a

4

.' a

BUM Whit U1.5MW U-'4 -aqUa WiUO'.SLMMO.Ma 4'UU'SCUUU 3 S-aeaulaub 4

aU

440'4444 -. a3544 .5.- Uha B hi 40.4W Wha�.4443 U0 '44'.. Wit -BCfINUCU �

aUg13

-'B- - �445.544 1 '404-U K WS%"

*OOONCUU..' ClU t'S4-UU

B,.5UUURUUK U

UKM U

UZISO oajU- -biM.U244...0S '4UZUKIU444b

U

O46 66

0660n...

* .... )0

41o.140.44414

, ,64 * 0 .

m.. ..... m66000000 4

06000400

*440 046 i

-- I• °1 44°°4 o 4

a66

66 40 06 0 O i l l l l l ~ a m m m m m ~ m

-- 79660 04

I4 . 4 NIN N U 4 . I- 1 - U

A A44 N M 1 MN4 AN2NItAAP4N ZE *44 M -4 "A 64

. . .. 4 .4 .. .- .4 .. . . .a

Li I S t l i t li t St s I &

A 11 NU6*1 mu....

4.~4 N 4NU N N t I It t tt l t it It I I

NO S4 U I* U Ul I Ul

*A 4 *Ut'*444.

N UN~iNno

* EU....A

u 4 I I 4 5 0 U 0 N 4 A U N

Ne Uv N A N A % 4 ~ f

2 - 4 -44. ; a4 ! .R .!614

444 0 R 9 UL GiOU MU U.. 00 0 N 44. NUNU U S N 4 - V.IL

LOUp ar Ot 4 4 UU UU N4N G - 080N

SL

u a

a MA 04 V P

Li m -000140

w Z AUN 44 AM

L, M i ll9 4a 4.4..

u a8 0.4944 944444 0444 994.99999904.999

3 *1.,949094400 S. .94494494 944U90 44 0494.... 1.4...... 141* .i9994494 9099 .4 4-94*.Non 9'904 .944 * .404496491

- 4944994940 N.44.9444. 04994 94 .94 94049449449499M491. U*49944444 04 44 944 *4 99 .94494099944440As4*

a M M W, me 4 143 ~ ~ ~ ~ ~ ~ ~ I 144 1! A9049994 .440 094.IU 9 z 4494494i04.0

* 0 , 4.9494%~~~~~ 4.004 4049 9P I~ 1991114

W 4:1 10 1 A 444449444444

4.4.4- 4011104 94.44049 4409044 09V.49

;00:0.494 00999 094.0 uI 411-111to141

40 ~ ~ ~ ~ ~ ~ ~ ~ I I . .I4499 .044 94 4 4400404944 44WE~ u044999 a,0 , 04 ,.94 4449040444.9.9

ai uu uu k 0.4%9.4494.o999

141090 w3499 094 .8 * :L11w.9999 w L40490 OIL449 a0494944u VI4,49409a

EE44'99499., .4099 . 04 994 94494944 4594999441

a at- & ý a90 osm16 -- s u a .. 1,a .

- LI 54 ULWX4 N

a w U u Ut. L .. Su .

a a 4 "mot lRiGU

U4~

4 ta PA to e H - Wew

::o INSSU J Ut

a Tt

asa

aa t$,.4 0 . 6 U 12 9NU~~tU~ a a~ 4~d.a~ U .t

1. L 110 6M, 4 NO;a .. *u: a

!tt..aUatt a -xwU .2aft H

ON a

a a a- -.tN~tt~aS ~A i. am 1

.4o~ l 'a . M U. ' -

&WaUL!&UUa 0 .- 5 ta 4

Ulali a US b Sa~ a82

A 0 61

00 .IAAAkiI 1 0 LiO 34

00a0 0 4. 400 UN

00 0 0.4 0401

me - 444.U ¶04.060004 0 .UU40..2

41 W* 'N 4U0 W .4404

00 0:. a1q "Is .8 66 4.I

4 mosses4. 4 ! 44 0lu

a4 .A 0W4 a , w,

4a a ts"0U 'A a44.q

4~~ U.UU0U0. 24446'6 U .I- aC C~ 16.1 C aL

Ii I Ii U, 0 aý U

2 itNO U U N 1 iO I 44000400 444U0 60 00 UUOP U

48 ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ 1 a&444 UN U .. 440 * U 0 . 00 604' O S S

a 44.40400644 40064 .0 044446 440 0 NE UUU .4 P IS

-Iq U6 4 -006U 44 a444~4f 4. 4 , , - .11U 4 004a 0UU aUa6a UU a 0 44 1W hi U44440: 9

000 a 0 .400. .*0 0.00 46 0 4 0 64

404 sa4.n0 .00 14 I"a116 1&

U 0 U 0 U

;:.0464 0 UN .4 44 sLCCLo a C 0U6 2444U44 0040C 04 U

a4P40 4 6UPIC

4 44111OUUOC 644vOua t 44 1 - o 03 U .hI

66C L UU 44 0 B .1 A.s44 . I, I.I 4444.4 aUUU.U44a56 6.4..400644 P464 44656 64

83~00 ~ ~

a aw a4

s- 4 a.-

U

48

S

Table C-2.

NASA - Lewis CET - 86

Output

Composition H

85

0

44444444444444

4444444444.4.4

.4444449444*44

44444444.4.444

4*4*444444.444

44444 44.4

* 4,... .4.4,, . .44

• 0~444

44444 4~4 E 444.

4. 4 4 4 4o4,

• 44 4444 44•

~4a*444444~444444444*44

UH CI W Q

.. ,CUNI.W&0L U

LN MS Mt qu put MSJ

a u 4* uI MU. U M L 0

I.,

I aw

limmm a a~

OUU6UUUUU4UUbULW

no~u an uua 0a

:HU 0.

at Nu u 000 0

I. *u I I.

Ai a a 0wme asa a a0

On 0 an W, 8. O5 a -

4, an uf a .5u~~vhr~0 a a . .8

888 .a . w r t c I.

ul a,.

amrn a a

urn.,,~~~ a a u r S M u . a a

ita a a aa 04 A

.a a W Un at a' a a. L U a I U'

.Suu ~ : a ao ma -. SQ 1187

o ,. fg~f aauw N .4 .0

3l#a aa

40W 0 a4 d, M4

333b3U :10144.

W UR4 30 a a 3S

sew& .*se auu e

.:"4441. 4*.AAZN "M *.AN a on oot .0f 40

*444041*4441*4 4.4. 44.4 . n%%%%N%.LN% %J',N lb. '4*%%Xa

U Z". 0. 4.

04 a 44

a clw

0.434

.3we au#4.a" a 4. .444

Zoo. .- 8 0 3.. 4# 44..*maumuma ue. ,.a 03aaa. aab.

I~~~I L44S .445 U E44.434 .I 4 .I *.444oau44muacuuaswaaauauuawss~~ u IwwW "1# *#5443

a Um 4 . k4. m ## 4 #4 4*%l.

441 #4#4 U #4U43455.40P4am"488

A

* . I.,

* II,.I, U

* I..- .4

U

UUU

* A

4

89

*4 14

30 01

.1 a a s

- 41 4 ,5 4 0 1 . N

- -. 0 5 1 10

44 0344......

F In

* 04

u L-~~~~~~~~~~~~~~ u44454 544444 6444*4-44.54t4a41

"!,N C 4*44166112* PC U0 L 4 l x 6 4U44P4NNCMNU44*444*4N44

* NP.#US* .A4vilS 04446 UP%#.4NN646449044

ft - W Wr%--C* ~ ~ -C t

a - 64 mwauaGzu ma w

-I L M U .6CCNCl;I'l - C a a . , 401

o u a Ce fftE.m.w A. !.CC - , UiC~l6.Lt6C'E UZa bw a a w u 2 3 3 m i-a-"- .- .. 4

Mm UCUU LC041UZ*UEU"

ai w

a4 a a C w wova

C Ut U C 3 M M 4 01 -U nNa u .6.ft4 IL t3. RC

a C uE I6 Z i A

&- - a maua- w U I 6.

aa.n w a.

* 6ulm

C C W91

a OR. Ne 1.

W a

aa aa n 0 a 0 %Aa

0aaaaaaa a * . . !Ra.. M1 .a a49im Ua 2 9xm ma a a 1u

20 0 R.69.ma amtawa~a~a 6.6 URUMNULRN'

A a 0

W Laa

w 0000 a, ma 049~ 00 11 1 00 .6aav! '

w 6*0 0. A . . .G m

V, ftNN~ x aOL UO

V, N4N9 a ama ONO& 6 -2"0* ~~ a 0r944am 4

* N Nmaoaaom a auat* NMo.. a. aNýaNG a

INf :aimaaaat

ON N C RO4.

* aaua

4- . No-

* aM N~m-.9. .064 N,9. U.mat. 2 aR9U ~ ~ ~ ~ ~ ~ . PON N-'sIN .. ..9 .a.U.SNN~

lo N Ni a a 2 a am

4 1. U u O 6 iw N 6.0 NN 9NNaa a a UN 0 Vi

I ~ ~ u~ a aiNNN~Na , aaam s 01 a

Z1 Iz UU a1410 a IS. N .1,:V. a.4.44l a a N-Wd mC4 . ufkk m. .9. 2 NN u I a x.91.

N N0~ilN ..a- Do92

0

.S Nf - A A. Wl

IV " I 4 I 0 " 4 I -0

aomm~fumat 4i

.4.

II uR N a.

L4 1* 4 : ý ý =SU

V .n ox4 ý! M

IxA. .4 1.

I.UE&V ft I.4 A

II~ftf~IUU93

---- a

x aa

on6 6; 6 a

144

m mm "a On .1rI

onamOn. a a a

Sa a an' m a

=" O

... N•

U4 f a0 a1 1

mamm1 a . 0 n

ame a

S..... a a* ua

I.mIe . 0 a"ace0 a a..... ... .... .. . *. a.., ..-- . .: "

tw .2 -. 4 C .* .4 O a a

-mmc a a

4 a

ENO aa A z : na00M

aS A A 0 a ott * -A

Nu 0 0 OCCaWOM M

10, 14 1 110 a, o

...S.. ...

WW~w6W .01 0 ma.0p SO

UNONIS M N w flVIVSS *

is 0 '5101010L.0&

S...95

it fla 4thmu

I

0 .n ac

-. mv Kinia ..

a~ ii

341 1 4 AU*

u3 U0aU. a.a

a a aa o'4Uv q 0.4 m

10 1

4 4 4 aaw w1

L1 N. aU It Wiafiff 4a1t3 6la 1 .. * LS a 4 w

U C n II I tmatahit.4t~i.,4aa~964

0 a. a wo ai u a

W .- Iu um A

-~~~ W A 4 2

-~. a 4 ., 4

fl UNNA W.4...WS .J 4w .J UWUa LIW 221S2 Um NM mo. a .2 0. 2.i 2 2

. .. . 1 h4m SAW noUWN a- .L

* 2WWi1 0 a 3w2 in I .IN2MOW u.4 '1. 1.6. .. 0 W U.1

I I x

4 1 W 2 L-

- I- W W2

4 W WI 2 4 97

C!.0 a~ C! aI a 44an n a 1! a 1"08 u

aenaana 0 a .mi..

E 6 4 :41

U2 44 K2

w 4 w a LI- 4 4 4 2 a

*I I a x t 200 ann U OI 3C4.

* .. a..aanM a 0jna a a U .u bai* UE.. IS *IONUEUc.o

& %abfa sIN.-Ldaui, CU

- I, au NaN..4aaq a

a ~ ~~~ w2aa4U f

16 a.., tipn t a% 4

I U amhlaaa a

-~~I 44 4 *.u

U~o' I. - ULJ.N&a a a nM iaa.u

4 44

"a Ona dA I

I i04 rMýV .. a*% 0a . a

13 d.a a.*.4 4N L~isf~sos pnN i slen 4 y 4 1

a4 nil0 * 'WU 1

& M46 .144~~ ~ ~ 4 iSau 4N 4 U . a 44 a

a s a O I4 W, .44 1. 4nana 4 10.IU b~I a Iba4~4 I5 I 4 aftIaa au. .

Is Un ua a s 5 U a 1 0am 46 a diuos~asa

3I U. l 4 saa2n. aaa ~ ~~~~~~ 4-9iN 0 1-n~n~ aN4 1 ~~ ...... 4a ~ ~~~ a I.P a.. qr amaaaa 4a o

0. 1 a1.. ,. 1E 1 L'n I u 4U.JM.d4 1

W 4'.~,N 3' -3 aW 41 a XUý'.41 u. 4'4 40 49 1O1-1.

-& W Aaa u &i la anna n

4 naa .1 * N 4U98

mt w

2 2 1 1 1

PI. I Z4 f

a al

wtgr n " aU 30at I 1M#MNUU

as .18

016 a w ,A

44 ~i In 0 14 " 4

u 02a a m

t t 41,

u ua aaa a a369 aa

6136103 Ufl499

Table C-3

NASA - Lewis CET - 86

Output

Composition L

0

0100

*449� 49944 94.9.

4�49� *994, 49949

* . 4.4.9

44�99 .. 9.. 9*4�99.... 9.... 9.9.99.9.9 *9999 9

49**9 999.99.,,, �44** 99.999..., �9�** 9�99999... �994994994 *999494944 .*949 49949944.. ***49 999999�99* *999* .99�49�49� 49994 94999949�* 99.9. 99,999499w 49999 94.99999*� *94** 9999999999 �9*49 4999999.99 *49999919* *�9*9 9.9.94..,, *9919 9999999499 *9499 �999*99499 9*999 9999494�99 49494 *949944.49 p999949*9� 44.49 4999499999 999*. 9.444

49494 �4949 9�99999944 *4*�999444 99**4 *999�49994 9�499 494�949499 *9999 99.9999449 9.949 4**99

44999 49949 4.49949*9* *4949 999*.49444 499�4 9499449�9� 94499 9499*

99949 *4�49494** 49**9 9490.99*94 9949* 49�949.499 ..... 44 9.9.99.444 94�49 4 4 4 9.*9�9.4.. *9949 4 4 S 9999494994 �99*9 4 4 4 999994949w *4.0. 4 4 4 4999949499 999.* 4 4 499499949, *44*� 999*944994 49944 94S994�4�9 94994 4444 4449949499 .9949 4 4 999�949999 99999 - 4 999944999k 99449 4 4 999994.... �9499 4 4 �494*99*4� 99494 4444 94,99999.9 *999999999 44949 94999

49444 *999� 4 9494494944 99999 4 9999949499 99949 4 4*944

49999 49494 444444 9999994949 49494 999*,99994 p99,9 99999

99499 *0994 99��94�99� *4944 444 99499994.. 94�49 544 99994.9,.. .4949 444 *9�99,999� 49�99 9999999499 49��4 9949999499 9..,.

49944 9444w 4 * 4 94999

994�4 999** 4 4 4 9�99999994 �9999 4 4 4 �494�9�444 ,4.9� 4 4 4 �.994� 999�9 4 444 �994994994 4�999 �999994994 9.94w 4 *9�99999d4 94499 4 �994�99499 99�99 * 449944,... *9949 = 99��9

99�49 99�9� 4999w99499 *4�9� 99999

449�9 99999 4 4 499�99999w 499.9 4 4 499.4

9. .9.9w 99949 4 - 99999* 94949 �9�99 4 * 49�99* 99994 �4449 44444* 94999- 99��4 p4499 99999

94999 9999k 49499- 99499 444444 99999 444444 9.99wo 99494 4 49994 4 49949- 99999 4 99994 4 9994994994 4 99999 4 49994.9 9�994 4 49494 4 999994 99999 444444 4999w 444444 99999* 9994. 99494 9494k* 9949w 44999 49994- 4q49. 444444 99.9. 4444 9999999999 S 99999 4 4 9�9990 ,99�4 4 �999� 4 4 44b94

99999 4 99��9 4 4 4.99499499 4 9499w 4 4 99999

99 444444 99�99 4444 49�9999994 9499k 94999

9*999ti 999�9 4 4 99999 4 4 499994 99499 5 4 9�9�9 4 4 4999a � 4 .. 44 4 * .. ,,,

999�4 4 4 4 4* 99999 4 4 99�99 4 4 99,9w

S9 4444 49449. 94�49 �9�99

101

iaa aa

:44ouLI.4.4U

*NNN NN4404

144

00 W00

0 0 0

N^ 00 a 000 -

a 00 00 .00

. U.4 u 043il.

U3 xN U .a U -6.U

UNUUUUUZUUML

N N102

LM4

allaa W a 9.f a .

4 L 4 1 8. 4 L

wino no*

. . . . . . .84h .4 . .4 o

8854 a* a .0.8.4 Na868O4m41a4~

!Sg.s*4* q a4.44 006CW R m .aatN 4.

I. U -. -o W,9 ws-w aK x R aa 0t l Is 88 -!400649 0000-~v - Q -"- U .

84 a3 8.83 fl -*- -OOSW4.8W8.8.83W LO, 4

4b.8I8.8L384NN"

atOU S -0 0 4 U 0 a -N .8 8 .m4 '.4I U I N N U .U8 N S N 8 4 4

ZUU~4.0 US8C .- N .40. 84. h0 ~U~i-Na m m aaamOn

I 50)8..036 W08UUm3..4'mm 4 II I 0w8.I fta .4U bS. .43 9uu maa~gm 406u uua8. .8884.8.8.888088Do

220 %NWS v.4N0m; a.

3 fl ý M N Ln S D.4444.4.4.4in . .3 .nII .4. .4.4 . .4 4 N

on .m m.mm

u a1

*0 . 4 a ! 0 1 0 4

3J LW i U W50 aW.400UM .. 8!8 4u Gmun zl'A'.md. aa

I ~ tU. 8-.8.'6' au *U8- ooba a.#4.4M8 0 a aaam m8.4EN.8b.8Um~~~~ !aoal, 6. u,. U4M~8S a a a a m m

asmu o.. A~wa~bm a~a war LI8.Ja~aEa - 4 ** *4* 4*

8.gM8U8 a Sr Iin.8t80E~aaaC.mnZ0 0 W0103000

* 4 4 4 ~ 1 4 . N4 t

I I I

u 44 4

w a ! .

U w 0 w a U a

a . 0 Via . a, .

L I L

u I 1 0

a " a. a

4l 4 4m a 6n - -

a i a in a 0 11 a R 'Ala w . . w . $4 a

a A T - a T 0 A0 u 'A i I 'A U I, u

3I w - UW40 61 I $4 $0

a4 a a4 2NEL NEEN

K4 A 2 4 0 94KNNE E

04 al U4. 61a a a u0 1413

we N4 N4 a44 0

UK uK 61 040

Cl 4 4 C 4 N A aCENN 1 1. . N 0 N UN

w U K2 N N N1 It4 N 4 z m W.a a 0 4,1 4$'31 U u w N as a N I I w k w w n a w Nt t t aNNU w 4 I NU I NI .NU I I U

a 0 a NW aI. NW0

L EL ELa 2, . m t DL

Q LS 'A 4 0 0 U0aI

E N

a U4 10 a a! u " . a a a ! a' N .1 aa ao 0, ao $4 ou ao am ts N aA $413$ a Clam

14 01 a1 U. N a" 43 14 u4.L 4 04 U .N N 4 0 U NA N L L 4 m u 4 K a

- - a r I I N $ I a a - N104a

a a'a a .

4 4 4 4

I~~ A.4 U 4 W.4L L N L4- C.

a 3 00

U W

a4 N1 4a.3

a .. .4 N .4 0 C 3

Nl a 1. N. N a 1 N N 3

* IU

40 a

*O a 1 44 "U O

w a "I w. 444 12 04 4 o- t

UU 0

a a u alo

p.I as'. -

k4 k it u a, lanno x x NIN

aA u I

04 A

a "no Oma as

*~~ 4, C, ui 5U

.b~go aab 4

A U u I U u I

LWj .444o WO W

-m a3 aa mNau%643 an W WO

41 .. . .

p. 44 mU a 4 RU un 4

40 2 a U a 42 4 w02 112 a 4 U a 0.. O uI a k0 a4 - 3is it a M, 0 t:1

q* w 45 9 .0 ý 4WO w6 0. p. 4, 1 04.

a 1 0 , , 0 z6 *. !.6

ma A a ; au u i to- m , a me auCCt s

6.44 LW 106

u I4

00

m

-A U I4. I 4

a L

It

o a

X Lmaa a

10LWU

mal

S !•.-..

.4 4. 4. 4 33• 4 ,

O aa 4.a

u s a eq a aq1

KNO

-. m US1 1 114 A I

Novommou Un .

3 II N 0"* 0 0p44

No 1,6 .14 O

a x 8

Do as

a 4 . . a o U

Z 44

- aha .U.4#43 dKK. ~ Is,4' SdCU6Ud4npSP#4 h #..K. .I# #.t 55.4 K., 55 .. PI. .4NP II I I 4Ip titS U p44.. # *-44b. *#0 4K SNi Ki'Sn" IKl I* 3 W U 4 U 4 K . a . S O K . , 4 . K .e ~ U .S;p . . 4 .#up 4 4

ha~~~~ ~~~~~~~~~ La f4 .44 NO OK4 #P a~IKCIKKC4~K#ha K K#K.# # K. . . .K4-pUP4 .p4 . K. p

1085v4# #K

u a

a IS A

*" W L, 6.

4. ~ ~ I .44 64

Is IN 4§ll N .

11 a .. E.E.U0L 0I IO 6 " TNl4 mo INXZE uWNLOINOO.4.m0!

* 4 WN I.PO II wa . UN N.l6S

11 L w ..

#41 11 1 o a . a w.

0. 68 asl a.s" u 0

.4 loft.WE.4..4

w -lw 0 low '66. Us 2" W

NIoEJEllN"WU4 . *U-

oI 4. N AU60.IUIEIU6UI

** mi iii

L u

*j W so ao so soo .. . .al U

109 *I W

5 Nl9 saO

In

V).. aa 'A 'A.

wpua z a U

W u a 4W

W a" a .a4

me13...: wassmoswo

Wom34 mo4 4

u Id a- a

S.1~ V aaa4aU~ 61 44

b B p.N40 a P3 a ti 11-~I a.' I.. a

W a p. Vp.U .u a I4 it MT 11

W 4-u mt o

own o a

VV- U U 4.4

W U U U UP-l

444 a a au ut4*WWW U al a, .g~W N~ a a~ Na N 80 t . . ,.

L A0s 1 4 4 6 umu'WUsNa .4 a.U U4

* ma. 44U44UN4110

S 6 aw W'

a U RS Ol o

Nau OUO

OS SS

0.6. It 9-U L W S w. V.

ULU: I

w~ I.

Owl"" N . -A

M.4 S A AI 1 4

NN U P..34 4..1 I8 aS N...u a 4.4

*ep"!m~e: NPSNU4NS Ia

no al

IN

a a.

U 4' 4

wUU U* a So u4~ 5 .

Table C-4

NASA - Lewis CET - 86

Output

Composition Q

112

4*444 44444 444*4

*444* *4*** 44444

4�4*q 444.4 0**4** 4 444**

4�4*4 *4*44 444�4

* *4,*4 44444* �4** . 4�***.4,., 44044 44044

4�4�4 4400. 404*44,... 44444

444*4 4,440 004*44*404 4444k 444,4

444�4 .4444 4�4*4

444�4 4,44. 4,444

44444 444�4 44,444�4*4 �40�� *�4444�44* 44444 4.444

4444* 44.44* *40� .44.. 4.44.

4�4*4 *444* 4*444

4�44a 44444 444,4

44444 *4444 4�444

44444 44444 444*440044 *04*4 44��44�444 .444. 4444444404 .4�0. .4,0.4�44I 444,4 4�444

44444 *444* 4444444444 44�4� 4444�444�4 .44., 44444

44404 444,, 444444�444 4444* 4�4444�444 44444 4.444

44444 *4*4* *44444�444 04444 *44444449g 444*4 *t 444�444444 *44*4 t t S *44�44444w *4444 � S 5 444444*4�4 44.4, � S 5 4�44444444 49444 t S S 4444444*�4 44444 S 5 444444�444 .4... 444�*44444 *4444 444444�4*4 44444 5555 *444444414 4444. 5 S 4*4*4

44444 � S 4444444444 44444 � S 444�*444�4 44444 5 S 444444*414 *44*4 5555 .4444

�444� 4.444

444.4 04444 555555 *4*�44*��4

I.,.. 44.4. 44.44* *044 *444* 4.44.

44444 *444444444 *4444 555 4....444�4 �4444 555 4*4�*

0�,�q 4*444 444444�44* 444444444w 44.4. 44444

4�44* * S 44444,,444 ..... S S S 44444

*4�44 44.44 555% 44�44* .4.4 0.*4� 4 44�04....

44�44 *4*4� = 5 *44*44444� S S � 4444- *44� p4444 5 5 00444

* *444** 4 *144 .444, .55555 e.*..- 0*0�� 44...

4,... *44�4- 4.,., *5*** 555555 90 440*4 S � s 4..,,.4,.,.. *04* 5

5 4,,,. S* ��444 S � 5* ,...4 5555tt � 555555 p4444

* 4.44 *4***.4 *�** 55555. 5555 44440

4�4*4 S 4444w S S 444.4* � . S 5 *44�4- 44444 5 4�4*� S 5 444�O

04.. S S 5 44444� 555555 4444. 5555

U 0 S 5 � S S* 4444 5 5 44444 5 4 4��44* 44.4 5 5 4 - S 4

p444 5 5 044�� 5 5 44�44* 4 5 5 . S S 4.4..* 44�44 SSS5 *4�44 5554 44444

- �4**� 444�4

S 113

1 11uw

04 N1.

u24 .422

14 UP m4 a 4

1a a 211

2 4.1 "

43 z* I

on o u t .a.

wG:Q~~b:":ma:4 a,. UWa'a wv

*00~o a1 no*4 22

ca.2C 6.l4 15o a o

: ~ ~ I toUa o co

mwaaommam'.sma 14 2.42

*UUUUU22UUUUNLW NOW442. a-

p.44 132114

a

4 aa a cc

LgUatfr4S.. NW L 0 t

mata OU an- ax; ,Z . tW *t4

It4 Ul taut 0 N a - Nu~abt-a6 tw400Nc etl0~

04 U 4- a. : 50 N. : ".!4 a . N I." I t Vt Vt I 4 N N N. 4 N .

MG AL JM J LL MM w vM M-M 44 4

* 3Nw a . a a

49 6 N N

a -0 noa a a4 OU X . 3 04X W. 6 4 N 4 4 6

CO. sun a a460614a6-

* L41N 3wa a -Sm--0 00

SON 4.1m4 a41414&NNNC ft-* C~ 36 mv404,4,6

a mtu 4mm

a a 4 1 -O a a u I .-

a bW :Im1-U 0.At0 30 u

W14 3 N N6 6 4414 to 1N4 "5 3 In a 5 40

ANN : !0 W" Ito 9.4 INý tai u N ~ 4 U 60N 4 4 1 4 4

L~ a a~-taua-a a aaaal i&m ain 606 00

0 ~ ~ ~~ a4v &000

4N~s~~r.t4 nN~in'UU NN N.44 O * m ~t1m15w

* a p. a N 4 4 . .N Nw N ftU . .4 N N 7 4.4.4.4 N l 4 4

* * a Ia N I 0* I l l N N N

.4 .4 .4

"ft ft P , 1 ft f . . ..U. .. .4 t N U S N f ft

a a

ai4U E16 4 a t f .4 .

00 N .1444%0L.414U

Us 0000080 004s00 es

- 4U 4141414140103

a~-0 i1400uoooooai a aa ~~

.4 O*N**N4m Aauc4m.4Dom..a a 4.44 U~uO aSUS u

K~~~ ~~~~~~~ .I. '4. 11 . . 4 . .11 . .441 .41*,41114

0 - '40wa 44 -4 - 0.*1R4'114411#41

w.4 .4 4 3 1..414 01449115..4.'11

cc0 .0 .11

* ~ ~ ~ ~ ~ ~ ~ ~ ~ 1 01 O41.1104 PU11 '4 04I1 9 1 , 1

a u I I00..3 0 .40 al u~ 0 -

U i 0441014 4 41 40 t a ".1 .14. 1- -3014* 114 *~a4.s 4 14 0 X.449 . 14 *..UUA L7 4u.n x 414..4414.44S44

0- - - -- - - - - - - - - -441.1 14 17 .O41 14 04 4 44 4

- a u W

I a. wN 4 u Wh

9 WNp 1.4was SAxN

N~hA~aLP .u a aa I a

a a

a ,

a .61 NoB 6C hl F. C I h

OGASl z w1 8:93iN

U LUN IN Whi IN*NNN

a x a

*l 0. a

* z a a a

a C a 4

0 .0 U S

hiu 10 hihihh h

X71ZON I hi*u4&.. uxm*ixa13

SU

a 8 a a 0 -

.C @4 aki.4a~a~a - looni

u M ws 0 0a 0 a 0 M w W 6 W

04 0 U

am a' am 'a on 40a I

- 0@*aads aaaw a w0. ai .. ; 0

44a a u a 0 a u a iS C.ath a 0a

* K0 4k 00V~~~a L4p.*.* a .4

UU U UIU N W40 0 6

BOi- hi; Isai.. .4. 0.40. aa

S; u u u a a Z sa a ww Iw *'*5f5a 1a* a.,a a W 0

A. w a x!68 U 0gW la al0 01.aa lu -.4r~a~~a *WargoC..

* U510S64'44CiA4 ***119.

a.U L2 4 9mA

L4 03 .4 44

a .414 I L O N .. 0

U- us

a. 00 ! ww 4 0

&3 N O&AE.R O& on 2.

4 4. 4.4 3 ..

044

444 0

i wi

~na4 4.4444 a 4a it

~~120

0 DISTRBUTION IJST ,No of Copics

CommanderU.S. Army Environmental Hygiene AgencyAMTN: HSHB-MO-AAberdeen Proving Ground, MD 21010

CommanderU.S. Army Environmental Hygiene AgencyAMTN; LibraryAberdeen Proving Ground, MD 21010

CommanderU.S. Army Missile CommandATTN; Propulsion Directorate,RDEC/Mr. L. B. ThornRedstone Arsenal, AL 35898

CommanderU.S. Army Medical Research and Development CommandAMTN: SGRD-PLCFort DetrickFrederick, MD 21702-5012

CommanderU.S. Army Medical Research and Development CommandAMTN: SGRD-RMI-SFort DetrickFrederick, MD 21702.5012

Defense Technical Information Center 2AMTN: DTIC.FDACCameron StationAlexandria, VA 22304-6145

Commander 2U.S. Army Biomedical Research and Development LaboratoryATTN: SGRD-UBZ-RA (Ms. Karen Fritz)Fort DetrickFrederick, MD 21702-5010

Commander 26U.S. Army Biomedical Research and Development LaboratoryAMI'N: SGRD-UBG-O (MAJ Young)Fort DetrickFrederick, MD 21701-5010

0 121

DISTRIBUTION LIST (Cont'd• No of Copies

Central Research LibraryBldg. 4500-N, MS-6286Oak Ridge National LaboratoryP. 0. Box 2008Oak Ridge, TN 37831.6286

Document Reference SectionBldg. 9711.1Oak Ridge National LaboratoryP. 0. Box 2009Oak Ridge, TN 37831

Laboratory RecordsBldg. 4500-N, MS.6285Oak Ridge National LaboratoryP. 0. Box 2008Oak Ridge, TN 37831.6285

ORNL Patent OfficeBldg. 4500-N, MS.6258Oak Ridge National LaboratoryP. 0. Box 62Oak Ridge, TN 37831

T. M. OayleBldg, 4500.S, MS-6120Oak Ridge National LaboratoryP. 0. Box 2008Oak Ridge, TN 37831.6120

M. R. Guerin I0Bldg. 4500.S, MS-6120Oak Ridge National LaboratoryP. 0. Box 2008Oak Ridge, TN 37831.6120

R. A. Jenkins 10Bldg. 4500-S, MS-6120Oak Ridge National LaboratoryP. 0, Box 2008Oak Ridge, TN 37831-6120

122 5

DISTRIBUTION LIST (Cont'd') No of Cool~

C. Y.MaIBldg, 4500-S, MS-6120Oak Ridge National LahoratoryP. 0, Box 2008Oak Ridge, TN 37831-6120

R. L MoodyIBechtel National Corp.151 Laayette DriveOak Ridge, TN 37830

C. W, Nestor 2Bldg. 9104.2, MS.8058Martin Marietta Energy SystemsP.O. Box 2009Oak Ridge, TN 37831-8058

Mr. J1. AX ReafanyderIEnergy Research and DevelopmentU.S, Department of Energy

Oak Ridge OperationsP. 0. Box 2008Oak Ridge, TN 37831-6269

C. V. ThompsonIBldg. 4500-S, IS-6120Oak Ridge National LaboratoryP. 0. Box 2008Oak Ridge, TN 37831-6120

B. A. TomkinsIDidg. 4500-S. MS-6120Oak Ridge National LaboratoryP. 0. Box 2008Oak Ridge, TN 37831-6120

Department of Energy 10Office of Scientific and Technical InformationP.O. Box 62Oak Ridge, TN 37831

123