P p -P

pp

P 1; i LA

p PA

pp ADpPpp

TECHNICAL REPORT ARBRL-TR-02476

SCANNING ELECTRON MICROSCOPE

EXAMINATION OF COTTON LINTERS AND WOOD

PULP FIBERS BEFORE AND AFTER NITRATION

AND GUN PROPELLANT MANUFACTURE

Donald C. Mann -T

Michael A. Patrick, A"4

March 1983

AARMY AMENT RESEARCH D DEVELOPMENT COMMANDBALLISTIC RESEARCH LABORATORY

ABERDEEN PROVING GROUND. MARYLAND

C...) Approved for public release; distribution unlimited.

LLJ

,--.

C.3F:/

P

Destroy this report when it is no longer needed.Do not return it to the originator.

Additional copies of this report may be obtainedfrom the National Technical Information Service,U. S. Department of Commerce, Springfield, Virginia22161.

The findings in this report are not to be construed asan official Department of the Army position, unlessso designated by other authorized documents.

T'hs us* Of trade~nk)AM Nr 'naz.ufaotumre I namea in this reportdoes not contituto i'idorsa~ir of dmni cownerial producat.

P

UNCLAS SIF IEDSECURITY CLASSIFICATION OF THIS PAGE tenw~ Doat Ent4w__

REPORT DOCUMENTATIOtN PAGE READ INSTRUCTIONSBEFORE COMPLETING FORM

1. REPORT NUMBER 2. GOVT ACCESSION NO. 3. RECIPIENT'S CATALOG HUMBER

TECHNICAL REPORT ARBRL-TR-024764. TITLE (and Subtitle) .5. TYPE OF REPORT & PERI3D COVERED

SCANNING ELECTRON MICROSCOPE EXAMINATION OF Technical ReportCOTTON LINTERS AND WOOD PULP FIBERS BEFORE ANDAFTER NITRATION AND GUN PROPELLANT MANUFACTURE G. PERFORMING ORO. REPORT NUMBER

7. AUTHOR(q) .. . CONTRACT OR GRANT NUMBEr,(s)

Donald C. Mann, Michael A. Patrick*

S. PERFORMING ORGANIZATION NAME AND ADDRESS 10. PROGRAM ELEMENT. PROJECT. TASKAREA & WORK UNIT NUMBERS

US Army Ballistic Research LaboratoryATTN: DRDAR-BLIAberdeen Proving Ground. XD 21005

11. CONTROLLING OFFICE NAME AND ADDtESS - 12, REPORT DATEUS Army Armament Research '& Development Command __..a 1983.

US Ariy. Ballistic Research Laboratory (DRDAR-BL) 13. NUMBER Or PAGES

4h Pdan Pr"v~vn~A mn . )I nn' 8414. MONITORING AGENC'V MANE I ADRESS((I dIfferang Ifoa. CmnfellhW Office) 1S. SECURITY CLASS. (of this report)

ISo. OC A31i PCATION/DOWNGRADING

16. DISTRIBUTION STATEMENT (o 0s1a Reporl)

Approved for public release, distribution unlimited.

I?. DISTRISUTION STATEMENT (of the abstrac entered In Blck 20. If dlffuani hem Repos)

IS. SUPPLEMENTARY NOTES

*US Air Force Armament Laboratory

Gun Rockets and Explosives DivisionEglin Air Force Base, FL 32542

11. KEY WORDS (Coillane an tootae aide DI necoeeamy and Ideutiiy by block aumber)

Morphology Gun Propellant Propellant MorphologyCellulose Propellant ManufactureNitrocellulose CombustionMechanical Properties Scanning Electron Microscope

2I0 A9U$TACr(Cmmt -iers e eb asouomy wli1g * 'by mock .ember) d31The prime ingredient in all operational military gun propellants isnitrocellulose (NC). The properties of the NC are a function of the sourceof the cellulose, any processes used to prepare the cellulose for nitration,and the nitration process. The purpose of this paper is to analyze themorphology of several standard types of cellulose feedstocks from a varietyof biological sources and preparative processes. The feedstocks examinedwere cotton linters, and sulfite or sulfate treated softwood and hardwood

DD = 147 FOWu e'no, OF 1 ov ss is os, .'vTi" 73 UNCLASS IF IED

SECURITY CLASSIFICATION OF T1hIC PAGE (When Data BEnefed)

P

UNCLASSIFIEDs5CUrTV CLASPICATION OF THInS PA6E13hm Daf t...

pulp fibers. Prior to nitration the cotton linters are well formed and intactwith smooth surfaces. The wood pulp fibers are broken and crushed from themechanical pulping processes with the surfaces of the fibers erroded andporous due to the purifying sulfate or sulfite processes. After nitration thesurfaces of the linters and pulp fibers show increased porosity anddegradation of the mechanical structure of the fibers with the cotton lintersshowing less damage than the pulp fibers. The increased porosity of the woodpulp fibers indicate why these fibers, after nitration, are generally moresoluble than nitrated linters. In both types of cellulose, the cell walls arestill intact with the microfibrils showing little damage. The cellulosemicrofibrils consist of long chains of the-polymer, bets, (1-4) D-glucosegrouped together to form bundles from .008 to .03 micrometer wide. Thesemicrofibrils'are held together by hydrogen bonds and appear as striations onthe surface of the fibers. The bundles of microfibrils are held togetherthrough covalent linkages by a variety of bonding matertals, including ligninin wood pulp, and form the inain structural element in the cell wall. Severalextruded single and double base propellants ,ere examined. The manufacturingprocess appears to destroy the mechanical structure of the cell wall whileleaving most of the microfibrils intact and lying in long strands parallel tothe direction of extrusion. in the' double base ball propellant WC 870 and thetriple base M30A1 samples analyzed, the microfibrils have been completelysolvated with no fibrous structure remaining. From these observations it isproposed that the total solvation of the microfibrils results in the formationof a brittle, crystalline NC matrix. This correlates well with the knownmechanical properties of fully solvated-NC propellants such as M30 andadvanced nitramine propellants using NC as the bidder. This suggests improvedmechanical properties can be obtained by retaining some of the microfibrilsand/or using plasticizers to prevent the formation of the brittle NC matrix.

UNCLASSIFIED

SECURITY CLASIFICATION OF THIS PAGE(thOa DO& 91t064

J w • m w m • * l

p

TABLE OF CONTENTS

Page

LIST OF ILLUSTRATIONS ................... .... o....o......5

LIST OF TABLES .....................................

1. INTRODUCTION ..................... .... ,.....,o,.....9

A. Cellulose .ie.............................................. 9

B. The Physical Structure of Cellulose.........................°11

C. Cellulose and Plant Cell Growth ..............................11

II. METHODS AND MATERIALS ................ 13

TII. RESULTS AND DISCUSSION ............. ... ........................ 17

A. Cellulose Feedstock Samples..°.............................17

1. Cotton Linter Celluloses............................. 17

2. Wood Pulp Cellulose.. ................ o................... 18

B. Nitrocellulose Fibers........................................18

1. Cotton itrocellulose....°........ ................... 18

2. Wood Pulp Nitrocellulose. ................ 6........ 19

C. Gun Propellants ..... ..... ..................... 20

.CONCLUSIONS ................ 00 .............. ...... 22

V. RECOMMENDATIONS. 666.6666 66 .......... 6 .66 666666666666.6666666666......24

ACKNOWLEDGMENTS. . .666666666666666666. 66666666666666.....66..... 24REFERENCES .... .... ................... * 666666 666 666666.6666 .6.6.6..78

DISTRIBUTION LIST ................................................ 79

L -41 ,e For,

* - 00d . ....

3 Dist Bea

P

LIST OF ILLUSTRATIONS

Figure Page

1. Cellulose [0 (1,4)D-Glucosepyranose] Molecule...................25

2. Model of Cellulose Microfibril ... S............. .............. .26

3. Model of Plant Cell Wall Bonding of Microfibril Bundles..........27

4. Model of Basic Cell Wall Structures of Cotton and Wood CellsShowing Laminar Structure of Cell Walls.................... .... 28

5. Buckeye Cotton Linter Cellulose Fibers ........................... 29

6. Hercules Hopewell Cotton Linter Cellulose Fibers...... ...... ,....31

7. Fractured Buckeye Cotton Cellulose Fibers........................33

8. ITT Rayonier Wood Pulp Cellulose: Fibers............,.......• .35

9. Fractured ITT Rayonier Wood Pulp Cellulose Fibers................39

10. Buckeye Sbuthern Pine Pulpwood Cellulose Fibers(Sulphate Processed). ....................... . ............... .42

11. Alaska Lumber and Paper Company Wood Pulp Cellulose Fibers. ...... 44

12. Fractured Wood Pulp Fibers ..................................... 46

13. International Paper Company Hardwood Wood Pulp Cellulose Fibers..47

14. Cotton Linter Nitrocellulose Fibers....................... ...... .49

15. Cotton Linter Nitrocellulose Fibers..................,...........52

16. Southern Pine Nitrocellulose Fibers..............................54

17. Softwood Pulp Nitrocellulose Fibers............ ................. 56

18. Wood Pulp Nitrocellulose Fibers ..... ........... ........

19. ]IR 4350 Single Base Extruded Propellant Grains..... ......... ...62

20. MR 4350 Single Base Grain Fractured Longitudinally..............64

21. IR 4350 Single Base Grain Fractured Laterally,..................66

22. MIO Single Base Extruded Grain Fractured Longitudinally..........68

P

Preceding page blank;

P

LIST OF ILLUSTRATIONS cont.

Figures Page

23. MIO Single Base Grain Fractured in Liquid Nitrogen .............. 70

24. HC25B Extruded Double Bsse ....................................... 72

25. M30A1 Extruded Triple Base Propellant............................73

26. WC870 Double Base Ball Propellant Grain..........................75

LIST OF TABLES

Table Page

1. Wood Pulp Impuritiesse ......................................... 1

2. Radford Cellulose Feedstock Samples..........,.........ooo,,,..13

3. Radford Nitrocellulose Samples........... .......... ............. 4

4. Finished Gun Propellant Morphological Samples .................... 15

7Precedig ...... ....

'Preceding page blank

I. INTRODUCTTON

The prime ingredient in almost all military gun propellants isnitrocellulose (NC). This material is used as either the major ingredient,as in single base .propellants which contain approximately 98 percent NC byweight, or as an energetic polymeric binder as in triple base and compositepropellants which.usually contain between 15 percent and 50 percent NC byweight. The HC is forrueu by ractLing celluloae uith a mixture of nitric andsulfuric acids. The cellulose used is obtained from either cotton (fibers)or wood pulp fibers. This cellulose is a biological material, and itsproperties as nitrocellulose will bi a function of its biological source,preparation processes, and nitration processes. The purpose of this paperis to analyze the morphology of several standard types bf cellulose feedstocks from a- variety of biological sources and preparative processes.These celluloses were then compared with a variety of nitrocelluloses madefrom similar cellulose feedstocks at the Radford Army.Ammunition Plant,Radford, Virginia. The morphological effects of the resultingnitrocellulose on processing into finished gun propellants was thenexamined. Since nitrocellulose is obtained from biological cellulosefeedstocks, the biological sources and growth processes must be examinedbefore we can describe and understand the morphology of cellulose prior toand after it is nitrated. The two prime cellulose feedstocks are wood pulpfibers and cotton linters.

A. Cellulose Fibers

Wood pulp fibers are the fibrous residue remaining after the mechanicalchipping of pulp wood followed by disintegration of the wood chips '4lth a

hot solution of ither calcium bisulfite (sulphite pulp) or sodium nydroxide(sulfate pulp).1 " These chemical processes are designed to resove thelignin and other impurities from the cellulose fibers. The sulphite processgenerally yields the purer cellulose fibers. High quality nitrocelluloseshould have no impurities remaining after nitra.tion.

The two main classes of trees yielding cellulose fibers are softwood orconiferous trees and hardwood or deciduous trees. The cellular structure ofboth types of trees is a highly complex subjgct, bt some general featuresare common to both types and will prove useful to examine.

The bulk of wood pulp cellulose comes from the xylem. The xylem isthat part of a tree which has ceased to grow and provides mechanical

1 F. Ait Hall, Wood ,Pup," Scientific Amerioan, pp. 52-62, April 1974.

2R. Norris Shreve and Joseph A. Brink, Jr.,Chemical P'rocee Industries, 4th

Edition, McGraw-Hill Book Co., New York, 1977, pp. 545-565.

Preceding page blank'

P

strength and conveys sap to the growing 'portions of the tree.3 ,4 The xylemprovides the main bulk of tree trunks and larger branches. The structureconsists of elongated parallel cells with cell walls that have beenthickened by the deposition of layers of cellulose and strengthened bylignin. Each wood cell is bound to the adjacent cells by an intercellularmaterial whose composition- is species dependent. Tn the celluloseextraction of wood, which consists of chemical action with mechanicalagitation, most of the lignin and other impurities are dissolved, and thefiber structure is disintegrated longitudinally. The result is a fibrousmaterial in the form of flat ribbons, usually with flattened internalcanals. The structure of the cell wall is somewhat porous due to the strongchemical attack and the removal of lignin. Table I lists the typicalimpurities found in wood. The fibers of coniferous wood pulp are theremains of cellular elements known.as tracheids which are elongated singlecells. In hardwood pulp, the fibers cotisist primarily of the sap-carryingvessels which were formed by the fusion of rows of shorter cells.

Table 1. Wood Pulp Impurities

O Lignin (20-301)

0 Hemicelluloses (10-2U%)

0 Pentosans (10-15%)

0 Waxes, Proteins, Fats (Trace)

Cotton fibers are the hairs or trichomes formed on the epidermis of thecotton seed? These trichomes are extruded from the seed during the first-stagzs of growth and are covered with part of the waxy cuticle of the.seed? The dried trichome is marked by very characteristic'twists orconvolutions that occur at irregular intervals along its length, producing aflat convoluted ribbon. The cuticle is extremely thin and is not foundafter the trichome has been nitrated. The structure of the cotton exteriorcell wall consists'of parallel bundles of spiral wrapped fibrils making arangle of about 300 to the axis of the fiber. (See Figures 1(a) and 6(a).)

The cotton linters used for nitration are the short (0.05-0.5 cm)trichomes remaining on the seed after the longer (2-10 cm) fibers have beenremoved. These cotton linters consist of two basic types. The first islike the longer fibers with a fairly wide and flattened central lumen orcanal and many convolutions. The second type consists of short, roundfibers tapering to a point with thick cell walls and a narrow lumen. Bothtypes of fibers are typically 10-1.5 Um wide. Acceptable nitration quality

Farnk 1). ifZee, Q luioge itr4Z, Inteaoience Publishers Inc., New York,1955.

4Myron ,. Ledbetter', and Keith R. PoteTr,,roduetion to the fine ,t, urctu'e ofM, Springer-Ver'Zag, New York, Heidelberg, Be.ln, pp.37-99, 1970.

P

fibers of both wood pulp and cotton cellulose consist of the remains of oneor more cell walls containing at least 95 percent cdllulose.

B. The Physical Structure of Cellulose

Cellulose is a rigid, linear, crystalline polymeric material consistingof beta, (1-4)-linked D-glucosepyranose molecules. (See Figure 1.) Theaverage chain length is determined by the structural purpose of thecellulose in the plant cell wall and the species of the plant. Cell wallcellnlose consists of highly crystalline and regular regions of purecellulose interspaced with small regions of irregular or amorphouscellulose. These amorphous regions usually contain small amounts ofimpurities. Present theory describes the beta-linked glucose chains in thecrystalline regions as forming flat ribbons with the glucose rings of eachchain lying .in the same plane while the hydroxyl' groups on the glucole,molecules protrude above and below the plane of the cellulose ribbon.J' (SeeFigures 2 and 3,) Bundles of these cellulose ribbons are calledmicrofibrils.' The microfibrils vary in width from 0.008 to 0.03' 'pm and areapproximately half as thick as they are wide? ,6 The microfibril bundlesare held together by hydrogen bonds formed between the protruding hydroxylgroups of the' flat glucose molecule. (See Figure 2.) The hydrogen bonding inthese microfibrils is very strong with the "microfibrils remaining intactexcept under severe processing conditions. Typical processing conditionsinclude elevated temperatures, strohg solvents, acids, and bases.

The main structural unit of cellulose in the plant wall consiets ofthese cellulose icrofibrils bonded together in a polymeric matrix much likefilament-wound rocket motor casings where the filaments are bonded togetherby polymeric resins. The microfibrils are covalently bonded together byvarious polymeric sugars and proteins? A model of this structure is seenin Figure 3. The covalent bonded microfibrils are often found grouped intofibrils. These celiulose fibrils are generally between 0.05-0.3 pm indiameter 'with an average diameter of approximately 0.15 pm and up to 20 cmor more in length. The* presence of the fibrils and their length, and theexact composition and amount of the fibril bonding materials are determinedprimarily by plant species and function, with'the microfibrile serving asthe main building unit of the fibrils. These fibrils then form the variousstructured features found in the plant cell wall.

C. Cellulose and Plant Cell Growth

Plant cells grow by first forming a primary oi ou er cell wall afterthe nucleus has divided during cell division (aitosisd This primary cellwall consists of layeie'd cellulose fibrils. ''In cotton linters, this primarywall consists of dense spiral wrapped layers of cellulose fibrils. In woodcells, this primary cell wall is 'formed of matted fibrils resembling matted

sPete,' Albeeheim, 1T1Ae Wala of..Growing Pant COiU," Scientific American,

pp. 81-95, April 1974.

6R. Stuart Ti peon, Editor, Adgqanea in Carbohydrate Chemistry/ andBiohemietM Academic Press, Vol. 26, pp.29?-349. New York, 1971.

11

felt. This primary cell wall is usually coated with an outer protective orbonding layer. Tn cotton fibers, this is a waxy cuticle to protect theexposed cell wall, as cotton fibers in the cotton ball consist of singlechains of plant cells connected end-to-end. In wood pulp fibers, theindividual rows of plant cells are coated by intercellular materials thatconnect adjoining cell walls together. Schematic drawings of typical cottonand wood cells are seen in Figure 4.

As the cell matures, a secondary cell wall (SI) is formed right next tothe interior of the primary cell wall. It is usually difficult todistinguish this initial secondary cell wall from the primary cell byoptical means. The dist n.ctive difference is that the S1 cell wall is madeof tightly wrapped layers of cellulose fibri.ls wound in a spiral pattern.Also, the S1 cell wall is usually better defined and more dense than theprimary cell wall. As the cell:. continues to mature, the middle layer of the

secondary cell wall (S2) is fotsed. This S2 wall is generaliy not as denseas the primary and S1 cell wall, but is..much thicker and provides the mainbulk of a plant cell wall. Cell wall growth is usually terminated by the

construction of a final and more dense inner wall (S ) that surrounds theprotoplasm of the young plant cell. As a general rule, each secondary cellwall is formed of alternating angles of'spiral wrapped cellulose fibrils.The angles of the spiral wrapping and the thickness of these layers aredetermined by the species, the maturity of the cell, and growth conditions.

As the plant matures to full size, the successive'layers of fibrils inmost celi walls are thought to slowly unwind and become more and moreparallel. In a mature cell wall,, the alternating angles of the cellulosefibrils are still present to provide structural integrity. The strength ofthese multipurpose structures is'.readily seen in these wood pulp xylem cellsas they support the entire weight of the tree while serving as transportvessels for the fluids found in the tree. While the basic cellulosestructure is the same in most plants, each species has its own uniquecharacteristics. Thus, cotton linters possess a highly defined, spiral-wrapped pattern of the fibrils while the fibrils in wood pulp fibers, whichconsists mainly of mature xylem cells, appear almost parallel. (See Figure4.)

As the plant cell reaches its maximum length and maturity, two mainprocesses occur. The first is the deposition of lignin. Lignin is acomplex polymer of condensed, substituted phenols. It is' water insoluableand highly r'esistant to chemical attack. Lignin is the major binder for thecellulose in the cell wall and provides the strength and rigidity 6f woodytissues. The amount of lignin deposited will control the flexibility of thecell wall. Other materials deposited during maturation involvehemicelluloses, pectin, and some proteins. This process occurs in bothtrees and in the woody portions of cotton plants. The cotton fibers containlittle lignin and are thus very flexible. The final step in the process isdrying up of the protoplasm of the cell and cessation of growth. The endresult is a plant cell with a rigid cell wall composed of elongatedcellulose fibrils surrounding a hollow center that is used to convey sap to

12

P

the growing" regions of the plant. In wood pulp fiber, the dry weight of thefibers is 40-60 percent cellulose prior to pro~elaing while cotton linterscontains 70-97 percent cellulose by dry weight. '

II. METHODS AND MATERIALS

This investigation was performed by analyzing scanning electronmicroscope (SEM) microphotographs of various cellulose feedstock samples,celluloses nitrated at the Radford Army Ammunition Plant, and finished gunpropellants. These samples included both cotton linters and wood pulpcellulose, nitrocelluloses made from cotton and wood pulp feedstocks, andfinished gun propellants. The samples are described in'Tables 2i 3, and 4.

Table 2. Radford Cellulose Feedstock Samples

RAAPType Cellulose Source Process Lot No.

1. Cotton Linter Hercules Hopewell --- HPC-5756

2. Cotton Linter Buckeye --- 5756

3. Wood Pulp ITT Ravonier Sulfite 5749

4. Wood Pulp ITT Ravonier :Sulfite 5749

5. Hardwood International V-9063(Dissolving Pulp) Paper Co.

6. Southern Pine Buckeye " Pre-Hydrolyzed ---Sulfate (96%Alpha Cellulose)

7. Wood Pulp Alaska Lumber & BleachedPaper Co. Sulfite

*Note: Dashes indicate data not available.

The cellulose and nitrocellulose samples were provided by the RadfordArmy Ammunition Plant and picked from stock available at' the time. Thesamples are of typical stocks on handand no effort was made to follow agiven lot through nitration to the finished propellant. Thus all thesamples examined (cellulose, nitrocellulose and propellants) are related toeach other only as being typical materials and products used in propellantmanufacture.

7E iZ Ott, H. M. Spu lin, and M. W. Grafflin, Editors, Cellulose and Cellulose

DArigativea Vol. 5, Inte'eoience Publishers Inc., Now York, 1954.8E. L. Akim, "CeLLuLose-BeLZWether or 07d Hat," Chem Tech, pp. 676-682,

November 1978.

13I'

*1-4

01ca U)l U) tn U) U) U) U) W) U)

14141 1414 141*1-4 9 r .p4 -A - 4q r

0 U) U1O4U ) 0 0 N V) b

004-4 0 4 N 4 N3 oI r4 44)H

V-4 to 4

000 0 d C )

to 4) 4i41 44 44

U3 Ln 0 0 0 0 Ln rj0

V-41 WtNV- - 00

o~- 04~ 4

01 11)0 0 U) 0.0N) 9-4 0 It ON4 CO 04 q CN V44 3 4

ONU t') t1 C4) .l4) 014 141141 144 - 4a 0 '0 ' U'sO' V-) U)-4 0

:j - 4?r-4 k4 kN-4- k00 .4 0r

V4 a000 0 0 t-0 0 0 ka co) 001 P 0 4 4 P.0 0 41 +j 12 4 t41

u 9 - F4 r . r4 ' . * *.r - 0 U) i

H I 3 .- 4 P4 0444 PA ~ NO Q.tU)044L p4 bOC

be r- N N N N %0 t- Ub o) N 4A 0 0~10 o "I in b

p- 1 4 . 14 (4 . 4 4 4 4 .14 > 4 44 IA

4) )4 I)4 to 4 4 .0 In) O

0 ~ ~ ~ ~ ~ ~ ~ ~ 1 P, r ,4 ~ .4 *p ~ r~.4 ~4 4 1 to ca

r~4 0 co M19~

14C '-4 b4 V .' ~ )

-4 CO M 0 4 I 0 m) ch 0K 14 4)

000 U)0 N4 C' 1N-44 -4) 4 44 :

4) V4 V C4) o L -4~ 1400) 11r 4a 40 a0u)

04~U L)u U, OO u u ) I t~b 1W.ri coto +J r- ~ 44

0 x )4)9..

. - N .P

Z

0 tAI.t~~O-4 It P419-

P-4 N v-4 4J 0)~ t 0 0

144

co4 C? 0

uO -d 0 Co.

0 at~ 0 0 0 In4I

0 4 -AJ V4 Ino ttl x 1.a

th 21 LLIE-

014 * 41

oc0 0n 0 0 In

0 .a P

4j-

00

0 * 0 -4

~ZI1~O00 - 0

*Xrr

In

V-1 L 0 1

IL W~

co 4.)4

S 0 "q 0 G *4 14 - 4

411 "I 14 *r $4 Z I

4.' :3 0 4.'

0 "-4 z z >, V-" 4 0) 0.r .0d 'w -4 > >. u 4j' *- r10 U LfIOp a bOa .0 r4 P4 %D 0 0 0 "

I ObN4 n l ( 4J 4.) j U 4. .$4g r 4 ' -4 *r4 . r 4

*r .4 . 4 0 0

P

The samples for 'SEM analysis were placed on collodial, graphite-coated,aluminum stubs and coated with 100 percent gold in an InternationalScientific Instrument PS-2 Sputter Coater. The linter and wood pulp fiberswere placed directly onto the stubs at room temperature prior to goldcoating. The liquid nitrogen fractured samples were prepared prior to stubmounting by soaking a small bundle of fibers in the liqid nitrogen. Thefrozen fibers were then immediately fractured with two pairs of forceps.The fractured fibers were then placed onto the aluminum stubs faceup.

The gun propellant samples were prepared by using a knife or razorblade to fracture the samples longitudinally (parallel to the direct'.on ofextrusion) and latitudinally (900 to the direction of extrusion) to providean end-on view of the sample. These samples were then placed faceup onto agraphite coated stub. The graphite paint was used to hold the sample inplace on the stub. The goal was to examine fractured surfaces rather thanthe cut surfaces of the propellant sample. The knife-cut regions tend tosmear and appear smooth and featureless under the SEN.

All the samples were then examined in an International ScientificInstrument Super III-A Scanning Electron Microscope. To avoid burning thesamples, a low acceleration voltage of 10 Kv was used along with minimumfocus/examination times, particularly at magnification greater than 500X.The prime method used was to focus the image very quickly and study theresultant microphotographs (micrographs) rather than the direct image on thevideo screen.

For size determination, each micrograph contains some portion or all ofa 55.5/mm sizing bar. The bar is made up of a 50/m length, a 0.5 um gap,and terminated with a 5/tim bar. Low magnifications will show the wholebar. High magnifications generally contain some of the end of the 50/

pm ar, all of the 0.5/pm gap and some or all of the terminal 5/pm bar.Thus, a constant reference scale is available independent of themagnification. The micrographs also contain the approximate magnificationof the original micrograph beneath the lower right corner and is not theactual magnification of the micrograph as printed in this report. Thesemagnifications are included to siuplify comparisons between the micrographsprinted in this report.

Caution must be exercised in evaluating the SEM data as any singlemicrograph or even a large number of micrographs of a single sample areadepicts only, the object or region viewed and not necessarily the wholesample. To obiAin a statistical sampling of even a small sample requires avery large number of micrographs. The field of microscopic gun propellantmorphology is relatively new and, while the micrographs contain tremendousamounts of information, this information has yet to be correlated toprocessing changes and combustion characteristics. The goal of this studywas to obtain a generalized view of the whole sample. The basic technique

used was to first obtain a number of representative samples of a lot, thento prepare a small portion of each representative sample. Thus, a large lotof~nitrocellulose was sampled in four or five places in the container; theneach sample was broken into three portions. A small sample of each portionwas then viewed under the SEM.

16

P

The sample in the SEM was first viewed at a low magnification to get ageneral view of the whole sample and to spot any regions that appearappreciably different. Each different area was noted for further'examination. If no regions appeared significantly different, then threedifferent regions on the sample were examined to obtain a good scan of thewhole sample. The sample was viewed at higher magnifications, looking atareas of interest.

This technique was used on the first sample. The second and thirdsamples of a given portion were similarily examined, not to map each sample,but to insure that a representative scan of the portion was beingperformed. Thus, if the second sample appeared similar to the first, onlya very few micrographs .were taken. Then, the third sample was examined tobe sure i-t was also.similar to the first and second samples. This techniquedoes not provide micrographs of every possible structure in a given sample,but it does allow a generalized study of the whole sample with the minimumexpenditure of time and expense.. Thus, .as'one grows more familiar with agiven sample .type, fewer micrographs are required and each micrograph iscompared against the existing data base. The main goal of this technique isto obtain a generalized, pictorial record of the sample for comparativeanalysis.

This report Includes micrographs typical of the samples examined ratherthan all the micrographs taken of each sample. The observations andinterpretations presented are based on a large number of micrograohs foreach sample and sample type. Interpretation of the SEM micrographs shouldbe considered as our best attempts at understanding the microscopicmorphology of these materials. These interpretations are subject to futuremodificati'ons as our understanding grows.. Comments and/or suggestions oninterpretating the results presented are'welcomed by the authors.

IMI. RESULTS AND DISCUSSION

Representative micrographs of the various samples examined are found inFigures 5 through 26. Each category of sample will be discussed as a groupwith specific micrographs being referenced'when'applicable.

A. Cellulose Feed Stock Samples

1. Cotton Linter Celluloses. There were no major differences betweenthe two cotton linter samples provided. (See Figures 5 through 9.) Thelinters were intact, showing little evidence of mechanical or chemicaldegradation. Chemical degradation is generally observed as partial or totalsolvation of the entire sample and/or various microstructures. Some fibersdid show some surface tearing of the outer cellulose layers as seen inFigure 6(d). Both linter samples show the typical mix of flattened,convoluted ribbons and the round, tapered fibers seen in optical micrographsof cotton linters. -The fiber surfaces appeared smooth and undamaged. Theoverall views of the samples showed few broken ends and/or short fragments.

The liquid titrogen frozen linters, as seen in Figure 7, fractured withfew cellulose strands, tufts and/or delaminations of the cell wallstructure. This suggests that the fiber was relatively "crystalline" whenfractured, resulting in a clean break. This also suggsts a relatively

17

P

uniform structure and pui-ity of the cell wall. Some delamination orsplintering of the fiber ends would be expected if the S2 cell wall layerscontained different mechanical structures and/or regions of differingchemical composition. This fracturing evidence alone is not conclusive ofstructural or chemical homogeneity.

2. .Woodpulp Cellulose.

a. Softwood Pulp (Sulfate and Sulfite Processed). The softwood pulpfibers (Figures 8 through 12) show evidence of both mechanical and chemicaldamage. The overall views of all these samples show many broken ends andshort fragments as compared to the long fibers found in the originaltrees. Most of the fibers exhibited hair-like structures on the surfaces,-giving evidence of the tearing apart of the wood chips that occurs duringprocessing. Figure 8(d) shows some of the chemical erosion that takes placeduring pulp processing. Here the chemical erosion appears as a partialsolvation or "eating away" of regions'of'the cell wall with irregularsurfaces and pits, but with generally smooth edges. The surfaces of thefibers appear roughened with many torn regions. Most of the fibers appearas flattened ribbons with the central lumen collapsed. The pits or porestypically seen in wood pulp cell walls can be seen in Figure 8(b) and8(d). There is no readily apparent difference between the sulfate andsulfite processed pulps.

The liquid nitrogen fractured fibers did not appear to break assmoothly as the cotton linter fibers. (See Figures 9 and 12.) Figure 12shows delamination of the secondary cell wall layer (S ) with the centrallumen readily visible. This suggests'that either the fbers warmed upbefore they were fractured or that structural and/or chemical differencesexist in these cell walls as compared to the other liquid nitrogen fractured

samples.

b. Hardwood Pulp Fibers. The hardwood pulp fibers, peen'in Figure 13,appear very similar to the softwood pulp fibers. 'The overall views of thesample show broken and torn fiber ends and numerous small, irregular fiberfragments. The main difference between the hardwood and softwood pulpfibers is that the hardwood fibers appear to show less mechanical damage tothe fiber surfaces. Some of the hardwood fiber surfaces appear to be assmooth as those of the cotton linters.

In summary, it appears as a generalized observation 'that cottonlinters are relatively undamaged while both kinds of wood pulp fibers showsigntficant mdahanical and chemical damage, wit the hardwood fibers being"less damage6 than the softwood fibers.

B. Nitrocellulose Fibers

There appear to be no major differences'between the cotton and woodpulp nitrocellulose linters. The cotton nitrocellulose fibers appear moremechanically intact than the wood pulp, but the difference is in degreerather than in kind.

1. Cotton Nitrocellulose. The cotton linter nitrocellulose samples(Figure 14 and 15) all appear very similar. The fibers show increased

'18

P

mechanical damage compared to the cellulose linters. The nitrocellulosefibers show regions where the cell wall is split along the fibril structuralpattern with the cellulose fibril bundles exposed. The surfaces shownumerous smooth-edged pores that suggests partial solvation with subsequentrapid drying. Partial solvation of the cellulose occurs during nitration,especially by nitric acid,which is a strong solvent for cellulose. Almostall nitration processes use one or more rapid drying steps.

The cracks in the cell walls expose the structure of the cellulosefibrils. One explanation for these cracks is that they are the result ofthe rapid drying of the solvent-swollen cellulose fibrils. This inducesstress on the cell wall structureand the cracks appear at weak points,cleavage planes, and at the end of fibril bundles. The fibrilspreferentially separate parallel to the 300 spiral twist of the fibrils seenin the cotton cellulose linter feedstock. This is seen in Figures 14(c) and15(d). The depth of these separations generally appears to be on the orderof I to 2 Um in most of the' samples. This gives an indication of the depthof the outer layers (S1 ) of the secondary cell wall. There was littleevidence in all of the samples viewed of any cracks or fractures transverseto the fibril bundles. This suggests that the cotton .linter fiber bundlesare more or less continuous throughout' the length of the cell wall and thatthe individual microfibrils are also very long relative to the length of thefiber.

2. Wood Pulp Nitrocellulose. All of the wood pulp nitrocellulosefibers appear very similar. These ate seen in Figures 16 through 18.The fibers appear to show more mechanical damage than the feedstock, woodpulp cellulose fibers. The nitrocellulose fibers have many broken and tornends and many short fragments. Many of these fibers appear to have beencrushed or severely compressed. The fiber surfaces have numerous cracks andsmooth-edged pores. These pores range from 0.05 to 0.3 Um in diameterwithan approximate average diameter of 0.1 um.

The wood pulp nitrocellulose fiber surfaces look very similar to thecotton nitrocellulose fiber surfaces, especially at magnifications above4000X. The main difference between the cotton and the wood nitrocellulosefiber surfaces is that the visible surface of the wood fibers appears toconsist of the primary cell wall layer while the cotton fibers surfacesconsist mainly of the outer (S1) secondary cell wall layer.

The stress crackings visible on the wood nitrocellulose fiber surfacesas seen in Figures 14 through 16 appear to be less regular than those ofcotton. The primary cell wall of wood is thicker than that of cotton, andis still intact in these nitrocellulose fibers. Sovra indication of theunderlying secondary cell wall structure can be see-a in Figures 17(c), (d),and (e) and in Figure 18(c). Figure 17(e) shows & region where the fibrilbundles appear to have been separated into a network of microfibrils. Thistype of network and the overall surface cracks of the wood pulp fibersindicate that there are very few structural fracture planes in the fibrilbundles and that the structure of the wood pulp primary wall is less-orderedthan that of the cotton fiber S1 cell wall. This is consistent with thebasic structure of plant cell walls previously discussed.

19

P

C. Gun Propellants

Figures 19 through 23 are of single base extruded gun propellants, IMR4350 and MIO. Figure 19 of INR 4350 single base propellant is typical ofhow extruded single, double, and triple base propellants fracturedlongitudinally (parallel to the direction of extrusion) and latitudinally(perpendicular to the direction of extrusion). The smooth areas on thefractured surfaces are where the knife cut the propellant grain. It can bereadily seen that the cut surfaces contain very little morphologicalinformation. These smooth regions are very similar to the kind of surfacesproduced by microtoming propellant grains.

The extruded single, double, and triple base grains fractured like alarge wood log. The grains fractured readily longitudinally. The lateral,end view fractures also split like a piece of wood with much splintering atthe fractured surface. One would expect a nonfibrous material to fracturelike a piece of ceramic rather than like a piece of wood. This behaviorsuggests that the extruded propellant grains are fibrous, that the fibersare oriented parallel to the direction of extrusion and that the fibers arerelatively long in relation to their diameter. Figures 19 through 24, indeed,show the fibrous nature of the nitrocellulose in these typical single basepropellants. The original grains were graphite glazed. Ve HIO propellantalso contains graphite in the formulation. Similar single base formulationswithout graphite in the formilation appear translucent, with color rangingfrom light green to orange prior to graphite coating.

Figure 23 is of an MIO single base propellant grain frozen in liquidnitrogen and then fractured longitudinally. Figure 23(a) shows twocellulose fibers still intact. The remaining micrographs of this figure arehigher magnification views of the lower fiber of Figure 23(a). Note thesimilarity of these micrographs, and, indeed, all of the higher magnificationviews of single and double base extruded propellants, and magnificritions ofthe cotton and woodpulp nitrocellulose fibers. A general rule is that thehigher the nitrogen content of nitrocellulose, the less soluble thatnitrocellulose will be. A possible explanation for the observed morphologyis that the basic structure appears to be a mat of shredded or unraveledfiber sheets of higher nitrogen content nitrocelluloses. These sheetsappear to be delaminated or unwrapped sections of the secondary cell wallstructure of the fiber. The two fibers seen in Figure 23(a) are partiallydelaminated.

The higher magnification micrographs of. Figure 18.through 24 appearvery similar to the high magnification views of the nitrocellulose fibers.These propellant micrographs show large.numbers of smooth-edged pores ofabout 0.1 Um in diameterand the laminar or sheet-like structure of theindividual secondary cell wall layers of the nitrated cellulose fibrils isintact. The fibril strands average 0.05 Um in diameter. This is typicalfor all of these samples viewed. In general, the basic microfibril or

fibrillar structure appears to be intact, with the mechanical mixina andsolvent systems causing delamination and unwrapping-of the primary andsecondary cell wall structure. There also occurs some separation of thelaminar fibrillar structure into small sheets or bundles of fibrils.

20

P

The HC25B extruded double base propellant seen in Figure 24 appearsvery similar.to the extruded single base propellant micrographs. The majordifference is that the fibrillar stucture is less distinct and thenitrocellulose appears to be more solvated. This could be due to thenitroglycerine plasticizer in the formulation. Plasticizers and solventscause nitrocellulose fIbrils to swell. This would make the fibrils lessdistinct. The nitrocellulose appears fibrous in nature and does notresemble gelatin,but rather a compressed mass of thin, fibrous flakes orsheets.

The H3OAI triple base propellant in Figure 25 appears markedlydifferent from the single and double base extruded propellants. Figures25(a) and (b) are longitudinal fractures at the outer edge of the grain.The graphite coating is readily visible at the top of these micrographs.Figures 25(c) and (d) are longitudinal fractured surfaces. Thenitroguanidine crystals are .oriented parallel to the direction of extrusionand appear roughly cylindrical in shape, with surfaces roughened with small.bumps and/or pits, The crystals appear to be separated from each other by athin layer of nonstructured binder. The nitrocellulose binder appears tobe totally solvatedwith all of the mtcrofibrillar structure destroyed.Note that'the Nd used in this propellant contains only 12.6 percentnitrogen. In all of the triple base micrographs Viewed, no evidence of thelaminar sheets of secondary cell wall or even loose agglomerations ofmicrofibrils, has been seen. The original cell wall structure of thenitrocellulose fibers appears to be totally solvated during the mixingprocess due to the strong solvents and the high percentage of nitroglycerineto nitrodellulose (approximately 40:60) in standard triple baseformulations. In the lateral fractures of Figures 25(c) and (d), thenitroguaidine crystals can be seen sticking out from the fracturesurface. The holes are the pockets in the binder left by crystals thatremained on the other half of the fractured pieces. Each crystal appears tobe held in place by a featureless, gelatinous binder with little porosity.

While each nitroguanidine crystal in all the samples viewed appeared tobe surrounded by the nitrocellulose binder system, not all the samplesdemonstrated the same amount of wetting by the binder. The micrographspresented in this report show the crystals to have an approximately 0.1'Umgap between the crystal surface and the binder over. most of the crystalsurface. Samples of a different M30 propellant showed no such gaps. Awell-wetted surface vi.th good adhesion between the binder and the crystalvould show no difference between the binder and the crystal, making it verydifficult to identify where the binder ends and the crystal surfacebegins. An SEM micrograph is a picture of the gold coating over the surfaceof the sample with the major differences seen being gaps (black areas) inthe gold coating due to surface features. The responses of the differentcherical species in a given sample to the electron beam is of secondaryimportance in this SEM effort.

The double base WC870 ball propellant seen in Figure 26 is made from amuch different process than the extruded propellants previously discussed.This propellant is made by totally solvating the nitrocellulose in ethylacetate. The micrographs show no fibrous structure in the propellantgrains. The grains fractured very cleanly,with no splintering andrelatively smooth fracture surfaces. Figure 26 shows only a porous

21

P

gelatinous type of material with no structure. The micrographs in thisfigure are all of the same grain. The overall view of the grain is inFigure 26(a). The coordinates on the side of this micrograph locate the

other micrographs in this figure. Note how the porosity varies over

different regions of the same grain.

TV. CONCLUSIONS

The results of this study indicate that the SEM does not as yet provideenough useful information to quantitatively discriminate between the fine

details of cellulose feedstocks and finished nitrocelluloses. Themicrographs are useful in providing a qualitative assessment of the

mechanical structure and quality of these fibers. This status is expected

to change as our understanding of the SEM data increases. The SEM analysisof the finished gun propellants shows great promise as a useful production

quality control tool. The micrographs readily provide information onthoroughness of mixing, solvation of ingredients, porosity (both at themacroscopic and microscopic levels), wetting and adhesion of binder to solid

particulates, measurement of grain dimensions, adhesion of inhibitor.

coatings, and overall product mechanical integrity. While the above is only

an initial list of potential useful SEM data, there is potential for evenmore useful data with an SEM coupled to an image analyzer where themicrographic data can be readily quantitized..

Prior to nitration, the cotton linters are well formed and intact with

smooth surfaces. The wood pulp fibers are broken and crushed from themechanical pulping processeswith the surfaces of the fibers eroded andporous due to the purifying sulfate or sulfite processes. After nitration,the surfaces of the linters and pulp fibers with the cotton linters showless damage than the pulp fibers. The increased porosity of the wood pulpfibers indicates why these fibers are generally more soluble than nitrate

linters. In both types of cellulose, the cell walls are still intact withthe microfibrils showing little damage.

The cellulose microfibrils consist of long chains of the polymer (B,(1-4) D-glucose), grouped together to form microfibril bundles from 0.008 to

0.03 um wide. These microfibrils are held together by hydrogen bonds andare very resistant to chemical attack. The nitrating conditions must beextreme to fully nitrate the cellulose; otherwise,only the surfaces on the

microfibrils will be nitrated. The microfibrils appear as striations on thesurfaces of the fibers. The bundles of microfibrils are held together

through covalent linkages by a variety of bonding materials, including

lignin in wood pulp. These bonding agents comprise the bulk of thecellulose impurities in all types of cellulose and are removed through thepreparative processes and during nitration.

IMR and MIO 'single base and HC25B double base extruded gun propellants

were analyzed. The manufacturing processes appear to have destroyed the

mechanical structure of the cell walls but most of the microfibrils areintact and are lying in long strands parallel to the direction of

extrusion. In the double base WC870 ball propellant and the triple base

M30AI samples analyzed, the microfibrils have been broken uA with theinternal morphology of the grains showing no fibrous nature. This

destruction of the microfibrils is attributed to the strong solvents used in

the manufacture of those propellants.

22

P

The relationship between morphology and mechanical properties appearsto be strong. The single and double base extruded grains were moredifficult to fracturs in any direction compared to extruded triple basegrains frectured in the same direction. Tn general, the single and doublebase grains were ore-resilient and less brittle than the triple base grainsof M.30 and M3OA1. All grains fractured easier parallel to the direction ofextrusion. The orientation of the fibrils in the single and double basegrains and the nitroguanidine crystals in the triple base grains wereparallel to the direction of extrusion in all samples.

The morphology of these extruded grains resembles composite structures,with the unsolvated fibrils and the nitroguanidine crystals acting as thefilament filler and the solvated-NC acting as tkte'matrix bindev. Thewetting of the unsolvated NC fibrils by the solvated NC appears good, andas both the fibrils and the matrix are the same material, one wouldreasonably expect good bonding to take place. The surfaces of theunsolvated NC fibrils should be partially softened during the mixing cycle,thus providing a good t'rface for bonding by the solvated NC. This does notappear to be the case with the M30 triple base propellants.

The NC appeared to be totally solvated in the triple base grains, withno fibrous structure remaining in the NC. The overall strength of a grainwill depend on the strength of the NC binder, the nitroguanidine crystals,and the adhesion between the binder and the crystal. The longituditalfractures of the triple base grains show much of the surface of thecrystals exposed. This suggests that the weak point in this system is thebinder-crystal bond. The binder regions between and to the sides of thenitroguanidine crystals showed tufts and strands of binder when fractured atambient temperature,with no evidence of tufts of binder adhering to thesurface of the crystals. This suggests that the binder fractured around thecrystals and that the fracture did not pass from the binder into thecrystals as there was no evidence of any crystal'fracturing in thelongitudinal fracture surfaces.

One explanation is that, even in the case of good wetting, the binderdoes not adhere strongly to the crystals. When samples are fracturedlaterally, the stress is on the crystals and the fracture is moredifficult. The nitroguanidine crystals appear to be somewhat flexible atroom temperatures. This is evidenced by such fracture surfaces showing manybent crystals (figure not included). This flexibility would provide thegrain with more resilience when stressed laterally. The poor adhesioninterface between the NC binder and the crystals could also provide readyfracture planes in the longitudinal direction, thus explaining the ease withwhich these grains will fracture in the longitudinal direction, both at roomtemperature and when frozen by liquid nitrogen.

From the above, it is proposed that this breakdown of the microfibrilsinto cellulose macromolecules by solvation and mechanical action duringprocessing results in the formation of a brittle crystalline NC matrix.This correlates well with the known mechanical properties of fully solvatedNC propellants such as M30 and advanced nitramine propellants using NC asthe matrix binder. Thts suggests improved mechanical properties of thebinder can be obtained by retaining some of the microfibrils and/or usingplasticizers or other polymers to prevent the formation of a brittle NCmatrix.

23

P

V. RECOMMENDATIONS

The data presented in this report suggest that much furthermorphological analysis is required of propellant ingredients, processing

variables, and finished products coupled to the performance results of the

finished formulations. The microporosity seen in the NC binders and thegaps around the triple base nitroguanidine crystals suggest further studieson the effects of pore size and crystal wetting. Thus, it is recommendedthat basic research be perforued to determine the tnitial pore diameters andthe depths that various flames will enter as a function of pressure. Theporosity also suggests studies on the mechanical and combustion propertiesof propellants as a function of pore size and binder-crystal adhesion overthe range of temperatures and pressure rise gradients gun propellantsexperience in operational use.

The wealth of data presented in a single micrograph requires a largedata base to successfully understand the information presented. It isrecommended that workers in SE14 analysis should first begin with the rawingredients and follow each step of the process with SEM analysis to morefully understand the SEN data seen in the finished product. We willpresently begin a study to follow a single lot of cellulose through each

step of the nitration process, and finally through propellant manufacture.We hope to correlate these data with the changes in viscosity, polymer chain.length, chemical composition, and ballistic performance in gun systems. Theultimate goal would be to perform this type of effort for each type ofnitration process and each type of gun propellant, both those presentlybeing produced and those under research and development.

ACKNOWLEDGMENTS

This eifort was initiated at the US Air Force Armament Laboratory,Eglin Air Force Base, Flori6a, by USAF Captains D. C. Mann and M. A.Patrick. All of the scanning electron microscope micrographs were taken byCapt. Patrick at Eglin AFB. The cellulose and nitrocellulose samples wereprovided by Mr. James Morris of Hercules Inc., Radford Army AmmunitionPlant, Radford, Virginia. The finalized interpretations of the datapresented in this report and the report were written at the US Army BallisticResearch Laboratory, Aberdeen Proving Ground, Maryland, by D. C. Mann afterseparation from Air Force active duty. The authors gratefully acknowledgethe support of both the Air Force Armament Laboratory and the Army BallisticResearch Laboratory, which has resulted in the publication of this report.

24

CH2OH OH CH2OH OH

-0 0 > 0---

OH CH2 0H OH CH2 0H

Figure 1. Cellulose [0(1,4) D-Glucosepyranose] Molecule

25

Figure 2. Model of Cellulose Microfibril (Reprinted from Scientific

American Inc. by permission)

26

CROSSLINKING POLYMERS 'pA

' CELLULOSE MICROFIBRILS

Figure 3. Model of Plant Cell Wall Bonding of Microfibril Bundles(Reprinted from Scientific American Inc. by permission)

27

LUMEN

SECONDARY WALLINNER LATER (33

SECONDARY WALLMIDDLE LAYER (82)

SECONDARY WALLOUTER LAYER 11)

INTERCELLULAR 6SUBSTANCE

PRIMARY WALL ANDSECONDARY WALL (51)

COTTON UNTER WOOD XYLEM CELLS

Figure 4. Model of Basic Cell Wall Structures of Cotton and WoodCells Showing Laminar Structure of Cell Walls

28

P .

(A).

375x

.B)', "m '... " .. S.

~A.t ..- ,

:.' * : . ..

150-1500xFigure 5. Buckeye Cotton Linter Cellulose Fibers

Comment: The surfaces of'the linters are relatively smooth and intact.Micrographs A & B show both the collapsed, convoluted ribbons' and theround linter fiber types.' The magnifications listed on all the photo-micrographs are from the original SEM micrographs. Due to changes of

29

(C) ,m , , , m

2250x

(D)

7SOOx

Figure 5. (Cont'd)

Couent (Cont'd): image size during reproduction these values no longergive the actual magnification. These numbers are included for referencewith'all of the photomicrographs of this report. See description ofmicrograph white size bars on page 16.

30

'.I

(A)

37Sx

(B) .....

Figure 6. Hercules Hopewell Cotton Linter Cellulose Fibers

Coimment: These linters show some mechanical damage. Micrograph Cshows the typical cotton cellulose striations at 300 and the surfacecracks at 600 to the fiber axis.

31

(C)

22SOx

(0) .

0-a ~ "

Figure 6. (Cont'd)

32

(A)

2500x

(B)

lie.

2500x

Figure 7. Fractured Buckeye Cotton Cellulose Fibers

Comment: These fibers were frozen in liquid nitrogen and then fractured.Notice that there is little evidence of major delaminations of the cellwall layers at the fracture surfaces.

33

(C)

3950x

Wiur . a ntd

34

P

(A)

(B)

ee,

375x

Figure 8. ITT Rayonier Wood Pulp Cellulose Fibers

Comment: These fibers are from Lot No. 5478 and were sulfite processed.Picture A is of an x-ray scan of these fibers. The large peak shows thepresence of sulfur in the fibers. Note the mechanical damage on the

35

(C)

150-1SOOx

(D)

Figure 8. (Contd)2500x

Comments (Cont'd): fibers as seen by the numerous hair-like projections(fibrils) on the fiber surfaces. Micrograph shows some possible chemicalerosion of the fiber surface around the pits or holes in fiber cell walls.

36

(E)

2400x

(F)

5600xFigure 8. (Cont'd)

Comment (Cont'd): Micrograph E shows the outer cell wall layer de-

laminated and folded back along the fiber surface.

37

.24'

5600x

Figure 8. (Cont'd)

(A)

1200x

(B)

1600x

Figure 9. Fractured ITT Rayonier Wood Pulp Cellulose Fibers

Comment: These fibers are from Lot No. 5749 and were sulfite processed.These fibers were frozen in liquid nitrogen and then fractured. Note thedelamination and splintering of the cell wall,'indicating possible chemicaland/or structural heterogeneity.

39

(C) -

2400x

(D)

Figure 9 .(Cont'd) 2400x

40

2400x

Figure 9. (Cont'ci)

41

(A)

375x

(B)

1O-1OSOOx

Figure 10. Buckeye Southern Pine Pulpwood Cellulose Fibers(Sulfate Processed)

Comment: This sample contains 96 percent alpha cellulose. Note themechanical damage on the fiber surfaces.

42

-I..

Figure* 10. (Cont'd)

43

(A) .

375x

WV!AA

150-1500x

Figure 11. Alaska Lumber and Paper CompafiyWood Pulp Cellulose Fibers

Comment: These fibers have been sulfate processed and bleached toincrease the purity of the cellulose. Note the large number of fiberfragments in micrograph A.

44

I.E t. ~ ....

(C)

2250x

Figure 11. (Contsd)

45

.n1

FT

-2400x

Figure 12. Fractured Wood Pulp Fibers

Commient: These sulfite processed fibers were frozen in liquid nitrogenand then fractured. Note the delamination of the secondary cell wallat the fracture surfaces.

46

(A)

375x

(B)

150-1500xFigure 13. International Paper Company Hardwood

Wood Pulp Cellulose Fibers

Comment: Most of these fibers appear as flattened ribbons with thecentral lumen collapsed. Note the evidence of structural damage to thefiber surfaces.

47

p

(C)

2250x

(D)

7500xFigure 13. (Cont'd)

48

(A)

160x

(B)

790x

Figure 14, Cotton Linter Nitrocellulose Fibers

Comment: These fibers are from Lot BL-3. This sample contains 11.7percent nitrogen. Note that the surfaces of the fibers are cracked andporous, with numerous, round, smooth-edged holes (approximately 0.1

49

(C) 72-1 iq+

NN.-

gv A..

3950X

(D).

Figure to4. (Ct% 'Y cComn Cntd: mconi imte) h es es ecnayclwall~ stutrVsvsbebeet .h rcsi tepiaycl al

50W t

(E)

~.dV~a.A

VA- AL:

dlL,

IN AP:

~ 4 -. Ux

FiSr 14.Cn'd

P

(A)

160x

(B)

790x

Figure 15. Cotton Linter Nitrocellulose Fibers

Con ent: These fibers are from Lot No. BL-7. The fibers show extensivemechanical damage and a large number of smail fiber fragments. (See micro-graph A.) Micrograph D readily shows the 300 spiral-wrapped fibrilbundles cracked and split apart.

52

3950x

Figure 15. (Cont'd)

S3

(A)

160x

(B) 4;

79OXFi gure 16. Southern Pine Nitrocellulose Fibers

Conmment: These fibers'are from Lot C-0936Y, The fibers show extensivemechanical damage with torn fiber ends and numerous small fragments.Note the bordered pits in micrographs C and D which are typical of

54

(C)

390-3900x

Figure 16. (Cont'd) M53x

Conmwent (Cont'd): softwood fibers. The patterns of the cracks aroundthe pits expose the structure of the fibrils around the pits.

(A)

e,.

.. ,,. , . . . 160x

(B)

• " .790x

Figure 17. Softwood Pulp Nitrocellulose Fibers

Comment: These fibers are a blend of Lots P-i and P-7. The fibers.display significant mechanical. damage and crushing. Note the po.rosityof the fiber surfaces in micrographs C and E. Both micrographs show

56

(C)

390-,3900x

(D) N.d4

~**~*'

Figure 17. (Cont'd)

Comment (Cont'd): the round, smooth-edged pores along with a largenumber of cracks in the outer Cell wall that expose the fibril structurein the inner cell wall layers.

57

(E)

11850x

Fi gure 17. (Cont'd)

58

P

(A)

160x

(B)

790x

Figure 18. Wood Pulp Nitrocellulose Fibers

Comment: These fibers are from Lot P-1 and show mechanical damage withtorn fiber ends, compressed fibers, and small fiber fragments.

59

(C)

39S0x

(D)

4000xFigure 18 .(Cont'd)

60

pIw

(E)1

t 4w

41.

?. -j t.S.e"e7# .1 * I- ?j4~r .f ,g n j

$61

(A)

30X

Figure 19. IMR 4350 Single Base Extruded Propellant Grains

Comment: These grains were fractured at room' temperature. MicrographA. shows a grain fractured longitudinally or parallel to the directionof extrusion: The upper right corner of the left grain in this micrographshows the region where the knife cut the grain during the fractureprocess. Micrograph B shows a grain fractured laterally or perpendicularto the direction of extrusion. Again, note the smooth region where theknife cut the grain.

62

(B)

50X

Figure 19. (Cont'd)

63)

P

(A)

56x

(B)

400xFigure 20. IMR 4350 Single Base Grain Fractured

Longitudinally

Comment: Micrograph D is a higher magnification view of the centerregion of micrograph C. Note that the fibers in the grain are orientatedparallel to the direction of extrusion, and parallel to the perforation.

64

1600x

8000x

Figure 20. (Cont'd)

(A)

80x

(B)

(B) r .. .

400x

Figure 21. IMR 4350 Single Base Grain Fractured Laterally

Comment: This grain was fractured at room temperature. The arrows indicatethe edge of outer grain surface which is graphite coated. Note the orienta-tion of the fibers.

66

400x

(0D)

1200x

F~igure 21. (Cont'd)

67

(A)

30X

(B)

1600XFigure 22. M1O Single Base Extruded Grain Fractured Longitudinally

Comment: This grain was fractured at room temperature. Note thesimilarity of the fibrous morphology with the IMR 4350 propellant.

68

(C)

8000x

Figure 22. (Cont'd)

69

(A)

(B)

" .800x

Figure 23. MID Single Base Grain Fractured in LiquidSNitoge

Conmment: This grain was fractured longitudinally while being immnersedin liquid nitrogen. Mlicrographs B, C, and D are higher magnification viewsof the intact fiber seen in the center region of micrograph A.

70

(C

5600x

(D) POP*

Figure 23. (Cont'd)

71

(A) r:' q~,: ..a.

.. . "

1200-3600x

(B)

II

• ..- qm .

1200-6000x

Figure 24." HC25B Extruded Double Base

Comment: These grains were fractured at room temperature. Micrograph Ais a longitudinal fracture with the arrow indicating both the directionof extrusion and the outer grain surface. Micrograph B is a lateralfracture with the arrow indicating the edge of the perforation.

72-

(A)

1200x

4000xFigure 25. M3OAl Extruded Triple Base Propellant

Commient: These samples were fractured at room temperature. MicrographsA and B are of the same longitudinally fractured grain surface. Thegraphite coated exterior surface is visible at the top of each micrograph.

73

(C) 7 --

Ami

J .7,

240-600x

Figure 25. (Cont'd)24x

Cormment: The nitroguanidine needle-like crystatls'are parallel to thedirectijon of extrusion whi-ch is indicated by the arrow. Micrographs Cand 0 are of tik different grains fractured laterally'.

74

1. 2

(A)

A

B

C

120x

(B) 2A

2400x

Figure 26. WC870 Double Base Ball Propellant Grain

Comment: All micrographs are of the same grain which was fractured atroom temperature. The coordinates on the upper right corner of eachmicrograph reference the coordinates on micrograph A. Note the wide

75

(C) i

2400-7200X

(D) 2C

Figure 26. (Cont'd) 2400-720OX

Comment (Cont'd): variety of morphologies present on this typical grain.Also note the lack of any fibrous structures in this propellant type.

76

(E) I B

1200x

A 5600x

Figure 26. (Cont'd)

77

P

REFERENCES

1. F. Keith Hall, "Wood Pulp," Scientific American, pp. 52-62, April 1974.

2. R. Norris Shreve and Joseph A. Brink, Jr., Chemical Process Industries,4th Edition,.McGraw-Pill Book Co., New York, 1977, pp. 545-565.

3. Frank D. Miles, Cellulose Nitrate, Interscience Publishers Inc., NewYork, 1955.

4. Myron C. Ledbetter, and Keith R. Porter, Introduction to the FineStructure of Plant Cells, Springer-Verlag, New York, Heidelberg, Berlin,pp. 37-99, 1970.

5. Peter Albersheim, 'The Walls of Growing Plant Cells," ScientificAmerican, pp. 81-95, April 1974.

6. R. Stuart Tipson, Editor, Advances in Carbohydrate Chemistry andBiochemistry, Academic Press, Vol. 26, pp. 297-349, New York, 1971.

7. Emil Ott, H; M. Spurlin, and M. W. Grafflin, Editors, Cellulose andCellulose Derivatives, Vol. 5, Interscience Publishers Inc., New York,1954.

8. E. L. Akim,' "Cellulose-Bellwether or Old Hat," Chem Tech, pp. 676-682,November' 1978.

78

P

DISTRIBUTION LIST

No. Of No. OfCopies Organization Copies Organization

12 Administrator I CommanderDefense Technical Info Center US Army Communications

ATTN: DTIC-DDA Research and DevelopmentCameron Station CommandAlexandria, VA 22314 ATTN: DRDCO-PPA-SA

Fort Monmouth, NJ 077032 Commander

US Army Materiel Development " Commanderand Readiness Command US Army Electronics Research

ATTN: DRCDMD-ST and Development CommandDRCDE-DW Technical Support Activity

5001 Eisenhower Avenue ATTN: DELSD-LAlexandria, VA 22333 Fort Monnouth. NJ 07703

I Commander CommanderUS Army-Armament liateriel US Army Ylissile CommandReadiness Command ATTN; DRSMI-R

ATTN: DRDAR-LEPnL Redstone Arsenal, AL 35898Rock Island,. IL 61299

2 CommanderCommander US Army Missile CommandUS Army Watervliet Arsenal ATTN: DRSMI-YDLATTN: SARWV-RD, R. Thierry DRSHII-RK. R. RhoadesWatervliet. NY 12189 Redstone Arsenal, AL 35898

Director CommnderUS Army MkAADCOM Heriet US Army Natick Research andWeapons Laboratory Development Command

ATTN: DRDAR-LCB-TL ATTN: DRDNA-DT, D. SielingWatervliet, NY 12189 Natick, MA 01762

Commander . I CommanderUS Army Aviation Research US Army Tahk Automotiveand Development Command Research and Development

ATTN: DRDAV-E Command4300 Goodfellow Blvd.. ATTN: DRSTA-TSLSt. Louis, NO 63120 Warren, MI 48090

Director 2 CommanderUS Army Air Mobility Research US Army Materials andAnd Development Laboratory Mechanics Research CenterAmes Research Center ATTN: DRXiR-ATLM1offett Field, CA 94035 Tech Library

Watertown, MA 02172Commander

US Army Armament Researchand Development Command

ATTN: DRDAR-TDCDover, NJ 07801

79

DISTRIBUTION LIST

No. Of No. Of

Copies Organization Copies Organization

Commander 2 CommanderUS Army Research Office Naval Surface Weapons CenterATTN: Tech Library ATTN: S. Jacobs/Code 240P. 0. Box 12211 Code 730Re.earch Triangle Park Silver Spring, MD 20910NC 27709

1 Commander9 Commander Naval Underwater Systems Ctr.

US Army Armament R&D Command Energy Conversion Dept.ATTN: DRDAR-TSS ATTN: CODE 5B331, R. S. Lazar

, DRDAR-LCA Newport, RI 02840S. WestleyD. Downs 2 CommanderS. Bernstein Naval Weapons CenterJ. Lannon ATTN: Code 388, R. L. DerrA. Beardell C. F. Price

DRDAR-LCE, R. Walker China Lake, CA 93555DRDAR-SCA, L. Stiefel

Dover, NJ 07801 1 SuperintendentDirector SNaval Postgraduate SchoolDeoDept. of MechanicalUS Army TRADOC Systems EngineeringAnalysis Activity ATTN: A. E. Fuhs

ATTN: ATAA-SL, Tech Lib Monterey, CA 93940White Sands Missile Range

NM 88002 3 CommanderNaval Ordnance Station

Chief Naval Research ATTN: P. Li StangATTN: Code 473, R. S. Miller C. Smith800 N. Quincy Street S. MitchellArlington, VA 22217 Indian Head, MD 20640

Commander Program ManagerNaval Sea Systems Command AFOSRATTN: SEA-62, J. urrin Directorate of AerospaceNational Center, Bldg. 2 SciencesRoom 6EO8 ATTN: L. H. Caveny.Washington, DC 20362 Boiling AFB, DC 20332

3 Commander 2 AFRPL (DYSC)Naval Surface Weapons. Center ATTN: D. GeorgeATTN: Code G33, J. L. East J. N. Levine

D. McClure Edwards AFB, CA 93523Code DX-21 Tech Lib

Dahlgren, VA 22448 10 AFATL/DLDLATTN: 0. K." Heiney

Eglin ArB, FL 32542

80

DISTRIBUTION LIST

N]o. Of No. OfCopies Organization Copies Organization

10 AFATL/DLV I Hercules, Inc.ATTN: M. A. Patrick' Allegheny BallisticsEglin AFB, FL 32542 Laboratory

ATTN: R. B. MillerAerojet Solid Propulsion Co. P. 0. Box 210ATTN: P. Micheli -umberland, MD 21501Sacramento, CA 95813

ARO, Inorporated .ercules, IncBacchus Works

Arnold, AFS ATTN: K. P. McCartyTN .17389 P. 0. Box 98

Atlantic Research Corporation Magna, UT 84044

ATTN: H. K. King I. Hercules, Inc.'5390 Cheorokee Avenue P4lin OperationsAlexandria, VA 22314 AVTL DLDL

ATTN: R. L. SimmonsEglin AFB, FL 32542

AVCO Everett Rach Lab DivATTN: D. Stickler I I TM2385 Revere Beach Parkway ATTN: H. J. KleinEverett,,MA 02149 10 W. 35th Street

Chicago, IL 60616Calspan CorporationATTN:, E. B. Fisher 1 Lawrence Livermore LaboratoryP. 0. Box 400 ATTN: M. S. L-355Buffalo, NY 1422S A. Buckingham

P. 0. Box 808Livermore, CA 94550

Foster Miller AssociatesATTN: A. Erickson 1 Olin Corporation135 Second Avenue Badger Army Ammunition PlantWaltham, MA 02154 ATTN: t. J. Thiede

General Applied Science Labs , WI 53913

ATTN: J. Erdos I Olin CorporationMerrick & Stewart Avenues Smokeless Powder OperationsWestbury Long Island ATTN: R. L.CookNY 11590 P. 0. Box 222

ST. Marks, FL 32355General Electric CompanyArmament Systems Dept. 1. Paul Cough 'Associates, Inc.ATTN:. M. J. Bulman' ATrN: P. S. GoughRoom 1311 P. 0. Box 1614

Lakeside Avenue Portsmouth, WH 03801Burlington, VT 05412

81

P

DISTRIBUTION LIST

No. Of No. OfCopies Organization Copies Organization

Physics .International 2 Thiokol Corporation27b0 ?Jerced Street Wasatch DivisionLeandro, A 945 77 ATTN: J. Peterson

Tech LibraryPrinceton Combusfton Research P. 0. Box 524Lab., Inc. Brigham City. UT 84302

ATTN: M. Sumerfield1041 US Highway One'North 2 United, TechnologiesPrinceton, NJ 08540 ATTN: K. Brown

Tech LibraryPulsepower Systems, Inc. P. 0. Box 358ATTN: L. C. Elmore Sunnyvale. CA 94086815 American StreetSan Carlos, CA 94070 Universal.Propulsion Company

ATTN: H. J. HcSpaddenRockwell International Black Canyon Stage 1, Box 1140Corporation, Phoenix, AZ 85029

Rocketdyne DivisionATTN: BA08 J. E..Flanagan 1 Battelle Memorial Institute6633 Cqnoga Avenue ATTN: Tech LibraryCanoga Park., CA 91304 505 King Avenue

Columbus. 011.432011 Science Applications, INC.-

ATTN: R. B.Edelman I 'Brigham Young University23146 Cumorah Crest Dept. of Chemical EngineeringWoodland Hills, CA 91364 ATTN: N. Beckstead

Scientific Research Assoc., Provo, UT 84601

Inc. 1 'California Institute of TechATTN: H. McDonald 204 Karman Lab'P. 0 Box 498 Main Stop 301-46Glastonbury. CT 06033 ATTN: F. E. C. Culick

1201 E. California StreetShock Hlydrbdynamics, Inc. Pasadena. CA 91125ATTN: W. H. Anderson4710-16 Vineland Avenue .California Institute of TechXorth'kollywood..CA 91602 Jet Propulsion Laboratory

4860 Oak Grove Drive2 Thiokol Corporation Pasadena, CA 91103

Huntsville ivision

ATTN: D. FlaniganTech'Library 1 Case Western Reserve

Huntsville, AL 35807 University

Division of AerospaceSciences

ATTN: J. TienCleveland, OH 44135

82

P

DSTRIiBUTJON LIST.

No. Of o. OfCopies Organization Copis Orsaizaton

4 Georgia Institute of Tech 1 stgers State UniversitySchool of Zrospace n g. Dept. of Mechanical andATTN: B. T. Zin Aerospace Engineering

U. C. Strahie ATTN: S. Temkin

R. Powers University 1hignts CampusAtlna GA 3New Brunswick, NJ 08903Atlanta, GA 30332 "

. 1 Rensselaer Polytechnic Inst.Institute of Gas TechnoloDy ... ]epartment of 11atheuaticsATTN: Do Gidapow .Troy, NY 12181

3424 S. State StreetChicago, TL 60616 1 SRI International

Propulsion Sciences DivisionJohns l pkins University ATTN: Tech LibraryApplied Physics Laboratory .333 Ravenswood AvenueChemical Proplsion Menlo Park, CA 94025Information Agency

ATTN: T. Christian .1 Stephen Institute ofJohns Hopkins l6ad TechnologyLaurel, MD 20707, Dividson Laboratory

' TTK: R. McAlevy, TITMaseh, ietts. Institute of Hoboken, NJ 07030Tech

Dept of Mechanical 1 University of CaliforniaEnAineer ng Los Alamos Scientific LabaT Tbrie Mn 02139ATTN: T. D." Butler 1S B216Cambridge, MA 01139 . P.O. Box 1663

Pennsylvania State University Los Alamos, 1.'M 85745Applied Research Lab .. 1 University of SouthknATTN: G. M. Faeth 'CaliforniaP. 0. Box 30 Mechanical Engineering Dept.State College, PA 16801 ATTN: 011200, M. Cerstein

Los Angeles * CA 90007Pennsylvania State UniversityDlpt. Of Mechanical 1 University of CaliforniaEngineering San Diego

ATTN: K. Kuo AMZS DepartmentUniversity Park, PA 16802 ATTN: F. William

Purdue University P. 0. Box 109

School of Mechanical La Jolla, CA 92037

Engineering University of nllnois1TTN: J. R. OshornTSPCChafee allDept of Nech EngineeringTSPC Chaffc all ecATTN: H, ErierWest itfayette, IN 47906.

144 MEB, 1206 W. Green St.Urbana, IL 61801

83,

P

DISTRIBUTION LIST

No. Of No. OfCopies Organization Copies Organization

I University of MiassachusettsDept. of Mechanica lEngineering- : "'.

ATTN' K. JakusAmherst, MA Oi002

I University of MinnesotaDept. of MechanicalEngineering

ATTN: 'E. FletcherMinneapolis, MN 55455

2 University of UtahDept. of Chemical EngineeringATTN: A. Baer

G. FlandroSalt Lake City, UT 84112

I Washington State UniversityDept. 6f"MechanicalEngineering

ATTN:- C.T. CrowePullman, WA .99163

Aberdeen Prov'ngGround

Dir, USAISAAATTN: DRXSY-D

DRXSY-MP,. H. CohenCdr, USATECOM

ATTN: VRSTE-TO-FSTEAP-14TA,

Dir, USACSLBldg. ES516, 9AATTN: DRDAR-CLB-PA

DRDAR-CLNDRDAR-CLJ-L

84

P

USER EVALUATION OF REPORT

Please take a few minutes to answer the questions below; tear outthis sheet, fold as indicated, staple or tape closed, and placein the mail. Your comments will provide us with information forimproving future reports.

1. BRL Report Number

2. Does this report satisfy a need? (Comment on purpose, relatedproject, or other area of interest for which report will be used.)

3. How, specifically, is the report being used? (Informationsource, design data or procedure, management procedure, source ofideas, etc.)

4. Has the information in this report led to any quantitativesavings as far as man-hours/contract dollars saved, operating costsavoided, efficiencies achieved, etc.? If so, please elaborate.

5. General Comments (Indicate what you think should be changed tomake this report and future reports of this type more responsiveto your needs, more usable, improve readability, etc.)

6. If you would like to be contacted by the personnel who preparedthis report to raise specific questions or discuss the topic,please fill in the following information.

Name:

Telephone Number:

.Organization Address:

p

FOLD HEREDirector NO POSTAGEUS Army Ballistic Research Laboratory 1PNECESSARYATTN: DRDAR-BLA-S IIF MAILEDAberdeen Proving Ground, MD 21005 IN THE

UNITED STATES

OFFICIAL BUSINESSPENALTY FOR PRIVATE U BUSINESS REPLY MAIL

FIRSM ,,CLASS PERMIT NO 12062 WASHINGTON,OC

POSTAGE WILL BE PAID BY DEPARTMENT OF THE ARMY

DirectorUS Army Ballistic Research LaboratoryATTN: DRDAR- BLA-SAberdeen Proving Ground, MD 21005

FOLD HERE