AD

"TECHNICAL REPORT ARBRL-TR-02308 V

Qi THE INFLUENCE OF PHYSICAL PROPERTIES

ON BLACK POWDER COMBUSTION

Ronald A. Sasse'

SDTIC
March 1981 SJUECTE JUN 1 6 1981,1

US ARMY ARMAMENT RESEARCH AND DEVELOPMENT COMMAND

BALLISTIC RESEARCH LABORATORY
ABERDEEN PROVING GROUND, MARYLAND

Approved for public release distribution unlimited.

8 6
[ ~81 6 05 O55N!,
 'p

Destroy this report when it is no longer needed.

Do not return it to the originator.

Secondary distribution of t'.is report by originating

or sponsoring activity is prohibited.

Additional copies of this report may be obtained

from the National Technical Information Service.
U.S. Department of Comerce, Springfield, Virginia
2216i.

The findings in this report are not to be construed as

an official Department of the Army position, unless
so designated by other authorized docunents.

Tlw u•wu ,on

tit" muwou or
wufado ,iaumoin thi'a iport
ouru'
.1o'p n.t *ivfliitute indov..niut of any oomwioraia prodAot.
 UNCLASSIFIED
SECURITY CLASSIFICATION Of THIS PAGE Mhen Data WnteMod

REPORT DOCUMENTATION PAGE U3TRUCTIONo

RZADCOMP--ZTW
BZFoRZ FORM
1' REPORT NUMBER 2. GOVT ACCESSION NO S. RMCIPIENT'S CATALOG NUMBERA

TEClINICAl, REPORT ARBRL-TR-02308 - r 3

47 TITLE (and Subtitle) S, TYPE OF REPORT & PERIOD COVERED

Tfil: INFIUENCIE OF PHYSICAL PROPERTIES ON BRL Technical Report

BLACK POWDE~R COMBUSTION____________________
C. PERFORMING ORG. REPORT
NUMBER

7. AUTHOR(W) 8. CONTRACT OR GRANT NUMBER(')

1
Ronald A. Sasse

MffOýZIt ? I A OD L 10. PROGRAM ELEMENT, PROJECT TASK

•JSi ,•Ii• a lJevciopment Command AREA & WORK UNIT NUMIERI
UhSArm.i' Ba lIi st i c Research Laboratory
A\TTN I)RDAR- B L Rl)'l11 11 (16 102A1143
:\berdeen Proving Ground, MI) 21005
II. CONTROLLING OFFICE NAME AND ADDRESS 12, REPORT DATE
US Army Armament Research & l)evelopment Command MARCH 1981
,LS .\Armv Ballistic Research Laboratory I. NUMBER OF PAGES
TFI'N: ' URDAR- B1I
Aberdeen Proving Ground. MD) 21005 fid
Il. MONITORING AGENCY NAME & ADDRESS(If dliferent from Controlllnl Office) IS. SECURITY CLASS. (of this report)

NI ,ASSq
ý I PIP.
Il
IS.. DECLASSIFICATION/DOWNGRADING
SCHEDULE

IS, DISTRIBUTION STATEMENT (of this Report)

Approved for public release; distribution unlimited

17. DISTRIBUTION STATEMENT (of the abstract entered In Black 20, It different liom Report)

IS. SUPPLEMENTARY NOTES

il. KEY WORDS (Continue on reveres aide If ncoee•iry mid Identify by block number)

Black powder
Propel lant
Physi cal properties of black powder

ý20 ANST1 ACT (Cauth eM venoshbNne...eemeavyan idsnty 67 block numober) (c it)

It has been observed that black powder made from ingredients of the same loi
and fabricated using the same manufacturing procedure results in material exhibi
ting different burning rates. The current study was undertaken to determine if
the physical properties of black powder affected burning rate and establish if
th•-se properties varied from one production batch to another.

(continued on back)
D JAN 71
II Jo" 1473 EDITION OF I MOV 49 It OBSOLETE INCLASS I F 11:1),
I SECURITY CLASSIFICATION OF THIS PAGE (kme Date Entered)

-. . I-
 UNCLASSIFIED
sacum"ITy CL.AUOPICATION Or ywgIs PA@Ug emet boav

20. ABSTRACT: (Continued)

Physical properties of black powder were measured including surface area,

pore size distribution, and internal free volume. Black powder derived from oak
and maple charcoals was compared, and the effects of processing the material by
a jet mill or ball mill were contrasted. Scanning electron microscopy revealed
that applied pressure, a manufacturing step used in making black powder, induced
considerable plastic flow that resultedi in a fused conglomerate, cohesive mass
containing interconnecting passageways of about 0.1 micron diameter. Density,
free volume, and surface area were found to be functions of burning rate. Based
on these relationships, the conclusion was offered that the degree of openness of
a grain of black powder is a significant factor affecting its burning rate.

i'
A¢
o::• ci•For

DI-I
I I

UNCLASSIFIED
SECURITY CLASilFICATION 0F THiS PAGE('.fen DOI&Enterod)
-1
 TABLE OF CONTENTS

Page
LIST OF ILLUSTRATIONS ............ ................... 5

I. INTRODUCTION .................. ....................... 7

II. EXPERIMENTAL ........................................ 12

A. Black Powder Samples ............. ................. 12
B. Scanning Electron Microscopy of Black Powder and
Charcoal .... ........... ............. 15
C. Compression of Sulfur, Potassium Nitrate and Charcoal 15
D. Mercury Intrusion Porosimetry ..... ............. .. 16
E. B.E.T. Surface Area Measurements .... ........... ... 17
F. Density Measurements. ................. 17/

III. RESULTS ................ .......................... ... 17

A. Black Powder and Charcoal S.E.M. Microphotographs . . 17
B. Compression of Sulfur, Potassium Nitrate, and Charcoal. 18
C. Mercury Intrusion FPorosimetry . . ........... 28
D. B.E.T. Surface Area Measurements ........... 28
E. Density Measurements ................ .............. 33

IV. DISCUSSION ................................. 33

A. Black Powder and Charcoal S.E.M. Measurements ... ..... 33
B. Compression of Sulfur, Potassium Nitrate and Charcoal . 35
C. Mercury Intrusion Porosimetry ..... ............. ... 38
D. B.E.T. Surface Area Measurements ....... ....... ... 40
E. Density Measurements .............................. 44

V. FUTURE PLANS ............................... .... ... 47

VI. SUMMARY ....................... .................. .. 49

ACKNOWLEDGEMENTS ................ .................. 49

REFERENCES ...................... .............. ...... 51

APPENDIX ............................................ 55

DISTRIBUTION LIST ................ ................. ... 59

3/.f
 LIST OF ILLUSTRATIONS

Figure Page

1. Comparison of burning rates as measured by flame spread,

0 (ref. 16); closed bomb, 03 (ref. 14); and open air
flamt spread, A (ref. 13) ...... ................ .. 14

2. Black Powder, Lot 1 . .. ................ 19

3. Black Powder, Lot 6 .. . . . ..... ................ 20

4. Black Powder, Lot 10 .................. 21

5. Black Powder, Lot 11 ...................... . 22

6. Oak and Maple Charcoals . . ............... 23

7. Oak and Maple Charcoals ......... . . . 24

8. Oak Charcoal Extracted from Lot 1 ............ 25

9. Maple Charcoal Extracted from Lot 11 .......... 26

10. Eiffect of pressure on potassium nitrate, sulfur, and

charcoal ........ ............. . ....... . . . . 27

11. Volume of mercury entering three samples of black powder-

Lot 6 .......... ..................... . . . . . 29

12. Volume of mercury entering three samples of black powder-

Lot 10 .............................. 30
13. Volume of mercury entering whole grains of black powder:
A-lot 11, burn rate O.60 m/see; *-lot 10, burn rate 0.51
m/sec; U-lot 1, burn rate 0.32 m/sec; and V-lot 6, burn
rate 0.38 in/see ... ............. . ........ 31

14. Volume of mercury entering half grains of black powder:

A-lot 11; burn rate 0.60 m/sec; 0-lot 10, burn rate 0.51
m/sec; O3-lot 1, burn rate 0.32 m/sec; and'*-lot 6, burn
rate 0.38 n/sec.... ........... ............ 32

15. Rclationship between burning rate and density of black

powder reproduced from reference 1. D~imensions refer to
pellet size ............ ...................... .... 37

16.
1 Relationship between flame spread rate and density.
Subscripts refer to deviant lot number .... ....... ... 39

* pF&Ch~jjjjU t.&Q 1LAL4KNOr FILAMiL

 LIST OF ILLUSTRATIONS (Cont'd)

Figure Page

17. Passageway diameter distribution for whole grains of black

powder. Dark portion repressed by a factor of 10 .. .. 41

18. Passageway diameter distribution for half grains of black

powder. Dark portion repressed by a factor of 10 .. .. 42
19. Relationship between flame spread rate and surface area.
Subscripts refer to lot number: 1-10 deviant lots; II,
75-2, 75-61, 75-44 are GOEX material; 7-11 is CIL material;
and M subscript samples were measured by Micrumeritics
Corp., others by NSW . . . . . ...... .. .............. 43
20. Rel,.tionship between flame spread rate and free volume.
Subscripts refer to lot number: +, values calculated by
equation (3) and , vwtlues calculated by equation (4) 46

21. Relationship between burning tir,.e and moisture content 48

10

2C

_J
j!
 I. INTRODUCTION

Black powder was made in production mills in Europe as early as

1340 in Augsburg, 1344 in Spandau and 1348 in Legnica. The process,
traceable to the early B.C. effort of the Chinese, evolved as an "art
form" having historical rather than scientific origins. The develop-
ment of the craft has been reconstructed by various authors and excel-
lent accounts have beeni given by Urbanski 1 and Marshall. 2 The subject
has been reviewed by Fedoroff and Sheffield 3 who provide numerous refer-
ences.

In essence, black powder has been made by grinding potassium nitrate,

sulfur and charcoal into a very fine powder. This material is trans-
formed into a cohesive conglomerate mass by either forming a paste which
is forced through and divided by a screen, or the material is pressed
into a cake. Both forms are subsequently broken into pieces. The small
lumps or grains of black powder are then coated with graphite to retard
the absorption of moisture and to prevent the grains from adhering.
The grains are then screened to a particular size. This process has
been standardized
nology, to some adds
but the operator degree, in accordance
water, with prevailing
adjusts temperature, selectstech-
grinding time, or changes compression pressure based on personal experi-
ence which is guided by individual judgment of color, odor, or general
p)erception of the condition of black powder as it undergoes its various
transformat ions. Such judgments are exchanged from experienced operator
to I ou-,1neyman.
Although the production of black powder was as great as 10.6 million
pounds 4 during World War I, the advent of smokeless powder diminished
demland. This fact together with the inherent dangers of production,
marked by occasional explosions, has resulted in closing most manufac-
turing facilities in North America. Because black powder still plays an
important role in fuses and ignition devices the U. S. Army Corps of
Engineers has built an Ammunition Plant in Charles':own, IN and equipment
was installed by ICI Americas, Inc. In designing the plant every effort
was made to modernize the black powder manufacturing process and to make
it as safe as possible. In the planning stage various studies were

1 TadeuvtUrbanaki, ,iemictry and Technology of Explosives. Vol 3,

Pergamon Preau, NY, pp. 322-340, (1967).

ýArthur Marshall, E.2ploaiveo Sec. Ed. Vol 1, Ilictory and Manufacture,

P. Bla•aton'o Son an( Co., Philadelphia, (1917).
3B. T. Fedorof'f and 0. E. Sheffield, •nyclopedia of E.xplosives and
Related It ems, Vol 2., PATR 2700, Picatinny Arsenal, Dover, NJ,
pp. R165-B178, (1962).
4 ArthuŽ' P. Vangelder and Hugo S hatter, Hiotory of the Explosive Industry
in America, NY Coliwmbia University Press, NY (1927).
conducted to: (1) elucidate and evaluate the ball and wheel mill process,
(2) to investigate new techniques and equipment and (3) to compare the
ballistic properties of black powder originating from this and other
countries. The accumulated knowledge was used as a foundation for the
design of the Indiana Plant. This work was reported in part by the
7
Chromalloy Corporation 5 Battelle Memorial Institute, 6 Olin Corporation
and IGI Americas, Inc. Battelle published abstracts of 50 articles
and 60 patents describing the wheel mill process in fine detail. This
process, sometimes referred to as the "standard process", and used in the
1860's can be currently witnessed at the Belin Powder Works in Moosic, PA.
The plant was built in 1911-12 and operated for many years by the
E. I. du Pont de Nemours tiCo. The plant is now owned and operated by
GOEX, Inc., and supplies most of the military black powder requirements.

Particle size distributions obtained by "the standard process" were

measured, using optical techniques, by Battelle Memorial Institute, 6 as a
function of both grinding time and moisture content. These results were
compared to black powder made by another technique, the jet mill process,
in which material is ground by impaction created by high velocity air
streams. The jet mill process was developed by Kjell Lovold 9 and is
used in Norway to make black powder. The technique was incorporated in
the Indiana Plant. Using the jet mill, particles are generally one third
smaller than produced by the wheel mill. Also, J. C. Alle:i 1 0 compared the

5 Chromalloy
Corporation, "A Study of Modernized Techniques for the
Manufacture of Black Powder", Propellex Chemical Division, Final Report
No. DAI-23-072-501-ORD-P-43, Jan 60, Chromalloy Corp., IL.
6H. E. Carlton, B. B. Bohrer, and H. Nack, '"atte Le Memorial Institute
Final Report on Advisory Serviced on Conceptual Deeign and Develop-
ment of New and Improved Processes for the Manufacture of Black Powder",
Olin Corporation, Indianz Army Ammunition Plant, 20 Oct 1970, Charles-
town, IN.
J. R. Pleseinger and L. W. Braniff, "Final Report on Development of
Improved Process for the Manufacture of Black Powder", RCS AMURE-109,
Olin Corporation, Indiana Army Ammunition Plant, 31 Dec 1971, Charles-
town, IN.
8 Indiana
Annun'tion Plant, "Black Powder Manufacturing Facility", Vol.
1 and 2, Indiana Army Ammunition Plant, 1975, Charlestown, IN.

9YKjell Lovold, U.S. Patent No. 3660546,

"Process for the Preparation of
Black Powder", 2 May 1972.
10
j. C. Allen, "The Adequacy of Military Specification MIL-P-223B to
Assure Functionally Reliable Black Powder", Report No. ASRSD-QA-A-P-60,
Ammunition System- ReZlability and Safety Div, Product Assurance Direc-
torate, June 1974, Picatinny Arsenal, Dover, NJ.

8
effects of different milling procedures at various production facilities
used in this and other countries.

Other types cf preparation have been attempted which include dis-

solving sulfurS by either carbon disulfido or alcohol and mixing this
solution, or slurry, with charcoal and potassium nitrate before evaporating.
Another innovation of note was pursued by VoigtII in which potassium
nitrate was precipitated from solution by adding alcohol in a controlled
manner to produce fine particles.
From the extensive background one general observation made in regard
to the production of black powder is that the same mabnufacturing proce-
dure, using the same ingredients, will result in lots exhibiting different
burning rates. This has been the experience of GOEX, Inc. where fast
and slow lots are blended to obtain a particular result. Although some
of the factors which effect burn rate are known, the lot-to-lot variations
have not been identified. One attempt to study the effect of small per-
turbations was supported by the Product Assurance D)irectorate, under the
direction of J. C. Allen1 2 where planned and slight deviations from a nor-
mal production run were purposefully and systematically introduced in
ten lots of black powder made by the Indiana pilot plant. These samples
were used to determine what manufacturing parameters are important in
effecting th performance and combustion of black powder. These lots
were termed - "deviant lots" and will be described later. The burning
characteristics of these and other standard lots were measured by four
different laboratories: ARRADCOM (BRL 1 3 , LCWSL 1 4 ), ICI Americas Corp. 1 5

aH. William Voigt, Jr., Pell W. Lawrence, and Jean P. Picard, "Process
for Preparing Modified Black Powder Pellets", U.S. Patent, 3937771,
Feb 10J 1976, see also H. W. Voigt and D. S. Downs, "Simplified Pro-
cessing of Black Powder and Pyrotechnical igniter Pills Including Low
Residue Versions", Seventh International Pyrotechnics Seminar, Vol. 2,
ITT Research Inst., Chicago I.L, July 2980.

J. C. Allen, "Concept Scope of Work For MMHTE Project 5764303 Accept-

ance of Continuously Produced Black Powder". Report No. SARPA-QA-X-10•,
November 1975, Picatinny Arsenal Dover, NJ.
13K. J. White, H. E. Holmes, J. R. Kelso Jr., A. W. Horst, and I. W. May,
"IIgnitionTrain and Laboratory Testing of Black Powder", ARRADCOCA-BRT,
Report No. IMR-64•, April 1979., APG, MD. See also K. J. White, H. E.
Holmes., and J. R. Kelso, "Effect of Black Powder Combustion on High and
Low Pressure Igniter Systems", 16th JANNAF Combustion Meeting, CPIA
Publication 308, Dec 1979.

' 4 L. Shulman, private communication, (data quoted in reference 13), 15

Jan 9 78, Picatinny Arsenal, Dover, NJ.

15 Hugh Fitler, private communication - Draft report, "Acceptance of

Continuously Produced Black Powder", 8 March 1979, XCI America Corp.,
Charlestown, IN.
 16_"

and Princeton Combustion Research Laboratories.16 These results were

compared by Kevin White 1 3 and will be presented later; it was concluded
that the same trends were observed in each laboratory. Although there
were differences in burning rates between various lots, the conclusion
was that small perturbations in manufacturing procedures do not account
for the difference of a factor of two in the observed flame spread rates.
When these lots were compared with some GOEX, Inc. black powder, it was
inferred that some other unidentified variable must be operative in
these samples.

The performance of black powder, in contrast to its preparation,

has been characteri.-ed in the classic papers of Noble and Abell7"18.
Recently, Williams summarized burning characteristics and included
several foreign references. Rose 20 studied the performance of black
powder inade from charcoal produced by various manufacturers and concluded
that response could b(, traced to volatile content; he also presented an
excellent review article. 2 1 Kirshenbaum"2 concluded from thermal DTA
analysis that even when volatiles were removed from charcoal the same
order of ignition was maintained and some other parameter besides vola-
tiles affected the ignition of black powder. lie also found that Ignition
temperature was not a function of carbon surface area, ash content nor
of sulfur content. Further, activation energy was not a function of
volatile content. Blackwood and Bowden 2 3 suggested that sulfur reacts
with volatiles as an initiation step but Kirshenbaum showed that all
three compononts, sulfur, potassium nitrate and charcoal, must be

l6Neale A. Messina, Larry S. Ingram and Martin Sunmerfield, '"ack

Powder Quality Asourance Flame Spread Tester", Final Report No.
PCRL-TR-78-101, Dee 1978, Princeton Combustion Laboratoriea Inc.,
Princeton NJ.

R. A. Noble and F. Abel, Phil. Trans. Roy. Soo.. London, Seriee A

Vol. 265, 49-155, (1875).
18
R. A. Noble and F. Abel, Phil. Trans. Roy. Soo., London, Series A
203-279, (2880).
29
F. WilZlczae, "The Role of Black Powder in Propelling Charges", Pica-
tinnyi Arsenal Tech. Report No. 4770, May 1975, Picatinny Arsenal,
Dover, NW.

James F. Rove, "Investigation on Black Powder and Charcoal", ITR-433,

Sept. 1975, Naval Ordnance Station, Indian Head, MD.
21
James E. Ro.c, "B l3ack Powder - A Modern Comnentary", Proc. of the 10th
Symposium on Explosives and Pyrotechnics, 14-16, 1979, Franklin
Research Tnot. Philadelphia, PA.
22
Abraham D. Kirshenbaum, Thermochimica Acta 18, 113, (1977).
A 23
J. D. Blackwood and F. P. Bowden, Proc. Roy. Soc., London, A213, 285,
(21952).
10

•:I
 present to induce a low temperature pre-ignition exotherm. Perhaps one
of the most persuasive experiments on the importance of volttiles was
that of Blackwood and Bowden in which they dissolved volatiles from char-
coal with acetone and applied them to charcoal that was 9S percent carbon.
Black powder made from this material had the same behavior as standard
powder.

Clearly, volatiles affect burning rate and Hintze 24 shows the rela-
tionship. lie recommends 82.5 percent carbon content for charcoal and
Blackwood and Bowden recommended 70 percent. However, neither work
relates these percentages to a standard initial form of water or ash-
free basis and it is not clear that the recommendations can be compared
directly. Both authors do recommend a rather high volatile content.
However, this parameter cannot be the only property that affects burning
rate for the deviant lots have diverse burning rates incorporating the
same charcoal.
In addition to the affect of volatiles, Shulman 2 5 ' 2 6 studied the

influence of small amounts of water, to two percent on ýhe burning rate

of black powder; this study was expanded by Plessinger. These investi-
gations quantify a well-known parameter.

From this discussion it is seen that the properties of charcoal are

important, but the exact nature of the requirements necessary to make a
good ballistic product are elusive. Present practice is that charcoal
be made from porous light wood, maple being currently selected, and be of
low ash content. No specification exists citing volatile content but
current practice is that it be 20-30%. This specification is broad in
that it requires charcoal which results in a proper performing black
powder. The specification does not cite the wood, particulars relating
to distillation, or physical properties. For black powder the amount
of potassium nitrate, sulfur, and charcoal is given but compaction pres-
sure is not. Only a density range for black powder is given. Such
criteria result in a large latitude in allowable variance in the materials
used to make black powder.

It appeared to this author that rather large variations in materials

or in preparation of black powder resulted in rather small but measurable

? 4 Woldomar Hintze, ExpLosivstoffe, 2, 41, (1968).

25L. Shulman, C. Lenchitz,L. Bottei, R. Young and P. Casiano, "An Analyseis

of the Role of Mloieture, Grain Size, and Surface Mlaze on the Combustion
and Functioning of Black Powder", Picatinny Arsenal Report No. 75-FR-
G-B-15, Oct. 1975, Picatinny Arsenal, Dover, NJ.
26
L. Shulman, R. Young, E. Hayes and C. Lenchitz, "The Ignition Charac-
teristics of Black Powder", Picatinny Arsenat TechnicaZ Report No. 1805,
Sept. 1967, F.catinny Areenal, Dover, NJ.

11A

:.1
N I
i'
variations in burning rate; lot-to-lot differences. using the sea.
materials and techniques, was a dominant feature. It was for these
re.ssons that another sensitive parameter was sought and a caidate hy-
pothasis was that the depee of openness of a grain could influence burn-
ing rate. Soon the study evolved into an investigation of the physical
properties of black powder and their relation to burning rate.
I
In an attempt to characterize black powder, scanning electron micro-
scope photographs were obtained which showed extensive fusing of material.
This immediately suggested that compaction studies be performed on pure
ingredierlts to determine what material flowed and the functional relation-
ship b,:cween flow and applied pressure. In addition the free volume was
measured by mercury intrusion techniques. Further, the internal surface
area was obtaird from the Brunauer, Emmett, and Teller (B.E.T.) gas
absorption technique. From these several measurements a physical descrip-
tion of black powder developed.

II. EXPERIMENTAL

A. Black Powder Samples

Deviant lots, mentioned earlier, were chosen for study as they were:
(1) well characterized ballistically by four laboratories, (2) made from
ingredients supplied by the same manufacturer, (3) fabricated by the jet
mill process, and (4) representative pilot plant samples typical of future
production techniques. Such samples were compared to black powder made
by the "standard" wheel mill process at GOEX, Inc. The deviant lots were
made from oak charcoal and GOEX, Inc., used maple. Equivalent compaction
pressures were used by both manufacturers.

Due to the limited capacity of the Indiana pilot plant, their jet-
milled meal was moisturized in a wheel mill at Picatinny Arsenal and
pressurized at GOEX, Inc., to not less than 3500 psi, or 246 kg/cm2 .
Fabrication has been described in detaillS and composition given in Table
1. For this table the carbon content has been recalculated on a moisture
and ash-free basis. Also to obtain the high and low carbon content char-
coals, three different lo'zs of oak charcoal were used. The high carbon
content charccal was but eight percent greater than that normally used
and the low carbon material was 17 percent below standard. The spread
in these values is significant but small. Such samples were compared to
GOEX, Inc., and CIL black powder produced by the standard process and
made %.tLh maple charcoal. The burning rate data from various laboratories1 6
ARRADCOiA (BRL, 1 3 LrWSL1 4), and Princeton Combustion Research Laboratories
are compared in Figure 1 reproduced from reference 13. The results are
normalized to the burning rate data for a standard GOEX, Inc., lot, 75-
44.

S. •I
12
 '.4 .4 L

Nul-4 0(a - 0
'j; a~O~
tE -0

-4P4 r4 P- - -

Lfl Uý Li n, 1*O 1ý
* '.d~ ~ t~ 0 ~ O N ~ In 0 qt

Nd 0..

tn -~4~4 4~

th. O N . N O E .

8' t* n0 * I N P-4~

0a 0 0

-4 -4 -4 - 4 - 4§L

ww

-4 M 0.1-t

u 0 0)
U sw. .4
-H 0
-~ 4)4)u

41" n D ' 0 m 1

0 00~'A0

13 ~. .
 160.

140
LUU
Z

41100- I

R
z1202
A- 804
I
0

0111 0 (ef 1I; lold olb,• r 14I); Idol r

o 2 4 6 8 10 12

Figure 1. Comparl son of burning rates as measured by flame spread,

0 (rf. 16); 1osed bombs (ref. 14); and open air
flame spread, (ref. 13).

14

iL
B. Scanning Electron Microscopy of Black Powder and Charcoal.

S.r.M. photographs of black powder were obtained at two magnifica-

tions, 300 and l000X, by John Gavel of the Analytical and Physical
Measurements Services Section, E.I. du Pont de Nemours & Co., Wilmington,
Di. Class one graini, those passing through a number 8 but not a number
4 sieve, were examined. They were cleaved in half by breaking the grains,
much like snapping a twig, to insure that no tool smeared the surface.
The new surface was coated with a 20 nm gold and palladium film; 30 keV
electrons were used for analysis.

Another imaging technique was attempted using an electron microprtbe,

model MS-64 made by the Acton Laboratories Inc., of Acton, MA. In these
experiments the -ample was moved normal to the oscillating beam by a motor
drive. The experiments were performed by Ralph Benck of the Penetration
Mechanics Branch (PMB) of BRL. Both sulfur and potassium x-ray maps were
obtained sNimultaneously using two detectors. Benck also analyzed the
charcoal ash, from charcoal extracted from lots 1 and 11.

Otber S.L.M. photographs of oak and maple charcoals were taken at

the BHR,by Gould Gibbons Jr., of the Explosive Effects Branch (EEB), using
an International Scientific Corp. S.E.M. model ISI-40 made ir.Avon, CT.
Oak charcoal was evaluated as received from the Indiana Ammunition Plant;
however, it was pre-screened. Charcoal passing through an 80 but held on
a 10t) mesh screen was examined. The maple charcoal was of a larger parti-
cle size paissing through a 10 but not a 20 mesh screen and, therefore, it
was ground in a mortar and pestle to obtain particles of the same size as
the oak. Both samples were obtained by the Indiana group from the Hardwood
Cha1c,1Cl Co. located in Steelville, MO. S.E.M, views were obtained at 100,
ISO, 500 an~d _00ORX.

In addition, o•ak charcoal was extracted from lot 1 and maple from
lot 11 black powder samples. They were extracted with water and carbon
disulfide. For lot 11 ,magnifications of 150, 350, 750, and 150OX were
employed; similar values were used for lot I and they were: 100, 250,
500, and 1000X.

C. ,ompression of Sulfur, Potassium Nitrate and Charcoal

Nureý sul'tir, potaissiumi nitrate, and charcoal were each ground to a

tfinc powde, usink, a mortar and pestle. Samples between 1 and 2 grams were
loaded into a I'vrkin filmer Corp. potassium bromide infrared die, model
180-0025, 1adC in Newark, CT. The powders were individually compressed
usi an Instron Corp. material testing instrument, model 2T'L[M, made in
Canton, MA. The press's cross heads move at various, constant, pre-
selcted rates ;uld the small compressive movement of 0.254 mm per minute
wa1s selected to insure some degree of pressure equilibrium. A total
load of 3,0t)0 kg cm- 2 tr 40,000 psi was applied slowly. At maximum
pressure the directionof the cross heads was reversed, at the same rate,
,• j

15

.4t
 to obtain a decompression curve. In some cases the process was repeated
twice to define the permanent effect of the first compression.

One sample of potassium nitrate was made damp by adding four percent
colored water and grinding to a constant hue. The experiments were per-
formed by Dominic Di Berardo of the PMB Branch of BRL.

D. Mercury Intrusion Porosit,.

"Internal volume and pore size distribution of either whole or cleaved

grains of black powder were measured by Jean Owens of Micromeritics Corp.
located in Norcro s, GA. The technique subjected samples to mercury pres-
sures to 3.5 X 103 kg cm'e or 50,000 psi forcing liquid into pores having
diaweters grvater than 30A. The grains of black powder were degassed by
a 1006C flowing nitrogen stream 2 for 40 minutes. The technique has been
described by Orr 2 7 and Adamson. 8

Mercury, due to its high surface tension, does not wet most surfaces
and pressure is required to force the fluid into small pores. Hence, the
pore size is directly related to applied pressure and the relationship
is given by equation 1.

Pr a -2 o cos 0 (1)

where P is pressure, r fs the radius, a is the surface tension of mercury

(taken as 474 dynes cm ) and 0 is the contact angle (taken as 130* for
most substances).

The distribution of such openings is given by a distribution func-

tion, D(r),

D(r) = (2)

where VT is the total penetraition volume and V Is the penetration volume

with openings greater than r.

.7Clyde Orr Jr., Powder Technol., 3, 117 (1969/1970).

" Arthur W. Adamson, Physical Chemistry of Surfaces 2nd Ed., Inter-
scienoe, NY, pp. 546-549, (1967).

Ih

1.
 E. B.E.T. Surface Area Measurements

In addition to the mercury intrusion measurements, Micromeritics

Corp. also determired surface area by a single point B.E.T. technique
using degassed whole grain samples and argon gas. This technique has
been described in several texts, for example Flood, 2 9 where surface
* area is related to the number of argon atoms required to form a liquid
film on a given surface. These measurements were extended to other
samples by Eleonore Kayser of the Naval Surface Weapons Laboratory, NSW,
in Silver Springs, MD, using the same type of Micromeritics equipment,
model 2205.

F. Density Measurements

liven though density is defined clearly as gm/cm , the measurement of

volume for powders and compressed masses can be ambiguous. For this
reason some experimental approaches will be discussed.

1. "Tap Density" is a term that has been used to describe the

weight of a powder -r objects poured into a known geometry and allowed
to settle by one or more gentle means such at "tapping". Such values
are not used in this study.

"2. "Bulk t)ensity" depicts the volume element inscribed by its outer
surface. The volume is measured by sinking a powder or object into mercury
at atmospheric pressure. Such values do not include pore volume and
are given in Table I as reported by Fit ler.

"3. "True I)ensity" is obtained by ineasuring the volume of vI object

as equal to the amount of non-absorbed gases it displaces. In such
measurements helium is usually introduced into an evacuated sample and
fills small pores. Similar values are obtained by using ether As the
displacement fluid for evacuated samples, 3 '( or using mercury dk:,placement
at very high pressures of mercury. In this study an ether disp',acement
measurement was made on non- vacuated samples to estimate the samples'
true density. Flow of ether into the sample was aided by placing the
pycnometer filled with black powL,ýr and ether in an ultrasonic bath.

Ill . RILSJLT,'s

A. Black 'owder and Charcoal S.Ii.NI NMicrojl 2 j'raips

Four bilack powder samples were chosen from the deviant lots for
examinat ion. Lots I and o arc samplus having slow flame spread rates of

'. Alvoon Plood, Ed., The Sol•d-Gao 1nterqaqcg Vol. 1, Marcel fDekker
Inc., NY, (1967).
3 R. M. McTnto•h and A. P. Otuart, can. J. Reo. B.4, 124, (1946).
 0.32 and 0.39 m/s as contr.sted to lots 10 and 11 which have faster
flame spread rates of 0.51 and 0.61 n/s. Further, lots 1, 6, and 10
are jet-milled oak products whereas lot 11 is a standard wheel-milled
material of GOEX, Inc. The S.E.M. microphotographs are given in Figures
2. 3, 4, and 5. Stereo views were made but add little to the detail
already shown and therefore are not presented. Potassium and sulfur
x-ray maps were obtained but the concentration profiles did not show
sharp boundaries as prominent features. The problem is the average
particle size is about 15 microns or less and the x-ray maps depicted an
average concentration to a depth of approximately 30 microns. Therefore
such analysis represented an average concentration throughout the slab.
For this reason those views have not been included.

The microprobe experiment was abandoned for black powder surface

analysis, for although the sulfur and potassium x-rays were recorded
simultaneously, and gave strong signals, the electron beam created a
reaction path and the data could not be relied upon to depict the
original sample. However, the microprobe did analyze the ash content
of both extracted charcoals and calcium was found to be the dominant
species and traces of silicon and phosphorous were noted. No other
metals were detected. The ash content of maple charcoal was 2.3 percent
and 11.9 percent was found for oak.

The S.U.M. microphotographs of the raw materials, oak and maple

charcoals, are given in Figures b and 7. Extracted charcoals from black
powder are shown for lot 8 in Figure 8 and for lot 11 in Figure 9.

B. Coýmression of Sulfur, Potassium Nitrate, and Charcoal

In the manufacture of black powder, damp meal typically containing

materials of less than 25 microns, is pressed into a cake where various
fabricators have used different compaction pressures. In general, a
pressure is selected which will result in a finished grain density of
approjimately 1.7 and GOEX, Inc. employs a pressure no smaller than 24(6
kg cm- or 3,500 psi. All of the S.E.M. microphotographs, Figures 2-5,
show considerable plastic flow and fusing among particles forming
extensive agglomeration larger than the original particle size. Therefore
a compaction experiment was performed to determine the relo'tionship
between flow and applied pressure.

The results of compressing pure ground sulfur, puossium nitrate

and charcoal are given in Figure 10. The relationships siown can be
linearized by presenting the logarithm of applied pressure is a function
of conalaction. The density of this sulfur cake was 2.00 gm/cm 3 and that
of th.- damp and dry l)OtSs1Sitm nitrate cakes were both 2.08 gm/cm3 .
Literatuire, values-"' are 2.07 and 2.109 gm/cm3 for t.he-,e compounds and
thus, at the maximum pressure used, no significant voids exist in eithe,'
material. In these cases the decompression curve is almost vertical
showing little elasticity and recompression forms a small envelope about
the decompression curve. The oak and maple charcoal compression curves

, 18
 Oil-

41~
'17
of-

L'"CJ

-mv--Aw.A9
I

0
4.
C
-I

*1
U
9-

M
a,
I.

5g
U-

o
0

*1

20
 qe~

Sim

Al WoU
 -i--

-J

.1

Go

22
--

CL
- -- -' - -I -i -
a m t ~~-,~-- ~---.--- -~-4 ~ -.

9C

Li.

140
 0',,

Li

4.,

I It,

LL
 ~CL

4J
JLJ

-1
 00

0 4-J

-J-

0 0

A-'

00
u

L4.

27
 were identical, but a compressed density of 1.01 gm/cm3 was measured for
the maple charcoal and a density of l.16 gm/cm 3 was measured for oak to
an accuracy of 0.5 percent. Values were obtained from the sample weight
and die dimensions at maxtumum pressure. The decompression curves for
both samples showed considerable elasticity. Interestingly, each sample
was removed as a powder from the die and obviously the organics did not
ict as a binder.

In an attempt to interpret the compression density vaiucs, the

densities of both charcoals were measured by an ether displacement technique
using a pycnometer. The density values were 0.99 and 1.17 gm/cm3 and
were accurate to 1.0 percent. They were equal to those measured in the
compression experiments. After compression, the charcoal originally
passing through a1i 80 hut not a 100 mesh screen, was agaI'in sieved and
the results are given in Table 2. The values show that many pieces of
charcoal were broken by the large applied force,

'rARLLU 2. CIIARCOAL S1IIVIVI AWI'l'lT COMPRESSION

Screeii Si:e Oak* Mapl e*

Mesh Units- 11

100 24 51
120 12 8
200 38 2b
340 26 is

Z by weight

C. Mercury Intrusion Porosimetry

Three different samples of 1-2 grains of whole grain black powder,

lots 6 and 10, were evaluated to determine the reproducibility of both
the sampling and measuring techniques. Data are given in Figure 11 and
12 showing the cumulative amount of mercury forced into pores of diff'erent
radii by various pressures. The d. A have been normalized at 200 psi or
14 kg/cm2 and shows a good degree of reproducibility. One sample each
of lots I and 11 was evaluated and this data is given together with a
representative penetration curve for lots I and 6 in Figure 13. These
same four lots were also evaluated by cutting the grains in half. These
data are given in Figure 14, In all cases the cut or whole grains were
intact after pressure was released.

:. .L.T. Surface Area Measurements

The surface area of whole grains was measured by 4 ,.".T. analysis.

All of the deviant lots were evaluated as well as GOEX, Inc. and CIL

}• " 28

ii ,I
 PORE DIAMETER (microns)
0 .06
1.0111, U• U
0.1 yi, ,"
•1 . . I.. U "
.01i ,
.002

.06
.05'

"• .04 '9

2
.03

U.02 9

.01 S 9'V

100 ainslO a aO ap0 api a 0000

a, ma ti_ nill, .I , I a i , 11i1
10100 '2 10000
*PRESSURiE |•,':2

9*,9

Fijure 11. Volume of inercury entering three samples of black pounder -

tLot 6

!2

.. .4
 PORE DIAMETER (microns)
1.0 0.1 .01 .002
t0

S0
.05

S °O
S.04-

*
- .C3 -
_z

,02-

010
+

1000 10-0-00 psi) 100000

100 A AIbi lk & &
I IAi- I
I I

100 1000 10000

10
PRESSURE (kg/cm2 )

black powder
Fioure 12. Volume of mercury entering three samples of
* Lot 10

30

" .4
 -~ - -D

LUQ
oLf
__ __ _ _ _ __ __ _ _ __ _ __ _ _ _ _ _C'

L* L

N ~

o snaII
N As:)sU
0
J4f:

o.am,
 CLJ

"if'
§i
p-j *

ticM

EJ

(8ew1 N-dSNIA~ a
 black powder. One lot, lot 6, was examined in detail to determine the
wvriance and precision associated with sampling and measurement. The
data are given in Table 3 and surface areas ranged from 0.5 to 1.0
m2 gm" 1 . Each individual sample was evaluated at least twice.

E. Density Measurements

The density of whole grains of black powder was measured by an ether

displacement technique and values are given in Table 3. They were
measured to an accuracy of one tenth of a percent.

The ether displacement density of charcoal extracted from lot 11 was

1.43. The densities of the maple and oak charcoals supplied from the
Hardwood Charcoal Co., were 1.01 and 1.16. Reproducibility was as noted
above.

IV. DISCUSSION

A. Black Powder and Charcoal S.E.M. Measurements

The microphotographs of deviant lots 1, 6, and 10, made by the jet

mill process, look much alike. They are shown in Figures 2-4. Each
I• figure shows two different views of a particular grain and these views
are shown at two different magnifications. The view at 300X was attempted
to get a general representation. A smaller subsection is shown at lOOOX
to provide greater detail. Higher magnifications were recorded but they
lose the general nature of the features shown. Sulfur and potassium
nitrate cannot be individually discerned; however from the relative abun-
dance most of the features represent potassium nitrate.

Figure 2 shows a large isolated charcoal particle; clearly, compres-

sion did not fill its pores nor did it crush the charcoal. Thus in this
instance incorporation of material into the charcoal pores, a process
that has been ascribed to the wheel mill process,S did not take place.
In all figures considerable plastic flow and particle deformation are
shown and many consolidated regions exist that are much larger than the
original particle size of 10-15 microns. Figure 4, deviant lot 10, shows
particles which appear more intact and are individually more discernable.
Also the degree of plastic flow seems less. This leads to a greater
degree of openness between particles.

The S.E.M. photograrh of the GOEX, Inc. ball milled material, lot
11, Figure 5, is difforet from the three-jet milled samples. Large
clumps of material are (,bserved as well as large void regions. Again
one piece of charcoal i•i evident and its pores are not filled, nor is it
crusned. The pores are :2.8 ± 0.9 microns wide. In general, material
seems to have flowed around the charcoal and the photograph shows almost
geological folding. Stress lines and spherical pockets are noted in

33
 TABLE 3. SURFACE AREA OF BLACK POWDER
AND ETHER DISPLACEMENT DENSITY
Surface Area** Densit
Surface Area*
m2 ipi-'I Cifl-
Lot Number m2 lm-1

0.50 O.588
1 0.588 1.91
0.53
2 0.73 1.87
0.80
0.54 1.88
3
0.58
4 0.060
0.060 1.89

5 0.58 1.85
0.58
0.5b 0.588
0.55 0.587 1.86
0.52
0.57
0.53
0.51
0.62
0.52
7 0.55 1.85
0.58
80.52 0.52 1.83
0.52
0.67 1.82
9
o.b3
0.bb 0.639
10 0.630 1.85
0.62
O.b6
GOIEX - Lot 11 0.82 0.801
0.76 0.796 2.02

GOKX 75-44 0.85

0.85
0.82
GOEX 75-b1 0.71
0.73

GOI-X 75-2 0.71

0.69
CIL 7-'.1 0.58
0.57

NSW
**Mio.mertice Corp.
At•erage io 0.55 ± 0.04.
34
 consolidated regions. The grain is not as homogeneous in appearance
as the jet-milled gr&in.

In summarizing the S.E.M. microphotographs, compaction seems to form

a fused conglomerate mixture laced with a maze of interconnecting passage-
ways.

The S.E.M. microphotographs of oak and maple charcoals are given in

Figures 6 and 7. Oak charcoal appears as separate lumps having a clearly
defined cell structure. The maple sample looks torn apart, perhaps by
the grinding action of the mortar and pestle. However, both samples
show extensive cell structure and broken and chipped pieces dominate.

One series of S.E.M. microphotographs of oak, Figure 8, which was

ground in the jet mill and extracted from lot 1 black powder, Figure 9,
looks similar to maple charcoal obtained in the same manner from lot 11
ground in a wheel mill. The particles seem tc be of similar size and the
only difference is the maple appears more rounded. No large charcoal
pieces are evident in the maple photograph as was seen in its black powder
counterpart but the view may not be totally representative.
In summary, it appears that grinding by either method produces a
fine powder which has a multiplicity of edges, broken pieces and con-
voluted surfaces. The grinding process breaks down the gross structure
of charcoal, forming pieces and fragments such that the two charcoals
look much alike.

B. Compression of Sulfur, Potassium Nitrate and Charcoal

Noting the extensive flow in grains it became desirable to quantify

the effect of pressure on black powder meal. Damp meal, that is generally
4 percent in moisture, has been pressed into a cake using various compac- 8
5 2
tion pressures. Values of 1900 psi (134 kg/cm ) to at least 3,500 3 psi
'
8,000 psi
(24b kg/cm2 ) 8 has been employed and some work was done at
2
(563 kg/cm ). The range of these values raised the questions, "do these
compaction pressures span the individual yield points of the ingredients
and is a particular pressure threshold required to induce plastic flow?"
To answer these questions an experiment was performed to determine if
the change in volume, or plastic flow, was uniform as the applied pres-
sure was increased.

The effect of pressure on the pure materials was determined and

shown in Figure 10. From the curvature, it can be seen that each
material flows throughout the applied pressure range. This is a
reminder that the applied pressure is a global value but the microscopic
pressure exerted between particles can be much greater. One compression

31 Memoriat Dee Poudree Et Salpetres, VoZ. 6, Chapter II by M. VisilZe,

Gaut'ier-Vallare Et Fil, 1mprfmeure-Librairje, Pa'ie, pp. 256-391, (1893).

£ 35

.- -
 curve using damp potassium nitrate, four percent moisture, lies well
under its dry counterpart showing that water reduces friction, allowing
* particles to slide. Thus, applied pressure and water content are coupled
phei.3menon which govern grain density.

It was noted that both potassium nitrate and sulfur formed smooth
glass like surfaces at the die faces. Such a surface could seal the pellet
and this effect could explain why black powder pelletized in a pellet press
exhibits slow relative quickness values.? Additional descriptions of
this effect can be found in the Ordnance Explosive Train Handbook 3 2 that
describes the burnring of a train of several pellets as a succession of
puffs caused by burning through the "glazed interface". In such appli-
cations, these effects are reduced by pressing pellets as interlocking
cones to minimize the effects of burning through a sharp boundary.

In addition to the surface effect of pressed pellets, there is a

density profile that should be considered when pressing to a given grain
density.

The charcoal compression curve is similar to that of potassium

nitrate and sulfur. From the facts that bulk density at maximum compres-
sion and ether displacement density are equal and that compression frac-
tures charcoal (Table 2), it is clear that pressure fractures some of
the charcoal into smaller fragments which then bend elastically to fill
the die leaving very little space between particles. With a density of
0.99 for maple
amorphous carbonandvalue
1.133 for
of oak
1.8 charcoal,
- 2.1, it both far removed
is concluded that from the
the density

of charcoal reflects an open structure.

The compression results show a gradual change in volume with applied

pressure. This reflects that applied pressure is a global value and the
associated microscopic pressure between small areas of two touching
particles is much greater. The smooth exponential curves show no sharp
transitions that could be associated with yield points nor are any
threshold values indicated.

The effect of compression on black powder meal is yet to be done but

the effect of pressure is to increase density. Vieille31 in 1893 demon-
strated the relationship between black powder bulk donsity and burn rate.
The data was reproduced by Urbanskil and his graph is given in Figure 15.
Obviously the data relate to a process and materials different from those
used today, but the trends should still persist. This same type of

3 Na7.aZ Ordnance Laboratory, Ordnance ltoaive Train Desianere Handbook-

* NOLR 1111, April 1962, Naval Ordnance Laboratory, Mite Oak, MD.

~Robert C. West, ed., Handbook of Chwnietvij and Phy eice,5ltdTe

Chemical Rubber Co., Cleveland., Ohio (2970-1972).

36

[-A
 100

10.5mm

80

"-g SLOW BURNING

IN PARALLEL LAYERS
LU 60

S403.5 m m

20

0 I I
16 1.7 1.8 1.9 2.0
DENSITY

Figure 15. Relationship between burning rate and density of black

powder reproduced from reference 1. Dimensions refer to pellet size.

317
relationship was shown to be operative in pyrotechnic mihtures where
burning rate was given as a function of compaction pressure. 3 4

Vieille investigated the large density neighborhood of 1.44 to 1.85.

In contrast, present experiments, using material that was available,
define a more restrictive neighborhood and the parameter is not illus-
trated as clearly. Since Vieillc's reference is not readily available,
the liberty has been taken to translate his summary. This translation
was undertaken by Eli Freedman of our laboratory (BRL-Applied Ballistics
Branch) and is given as an appendix.

'The flame spread rate and densities for the deviant lots, given in
Table I are shown as a scattered presentation in Figure 16. A functional
relationship appears to exist but its exact nature is not given by these
limited data. 'The scatter leads to the belief that compression condi-
tions were not identical; however, it is clear that flame spread rate
decreases as density increases.

C. Mercury Intrusion 1Porosimetry

Equation (1) relating pore size and applied pressure presumes that
the pores are a collection of right cylinders of different radii. In
practice small openings exist leading to a larger volume element. Such
a situation has been termed an "ink bottle effect" and when encountered
will lead to an overestimate of the number of pores that exist at a
particular radius. For loose powders, corrections are attempted by
measuring the mercury ,xtruded as a sample is depressurized. In accor-
dance with equation (1) the neck of the "ink bottle" will empty as well
as the "bottle" itself but at different pressures. For compacted
material, such a black powder equilibrium may be too slow to attempt this
correction depressurizing technique. These problems should be considered
when examining the intrusion data and the indicated radii should be con-
sidered as approximate values. However, the mercury volume entering a
grain is exact.
In examining the S.E.M. microphotographs of black powder many voids
are apparent with radii of 10 microns or less. For mercury to pene-
trate the grain and reach these voids it must intrude through passage-
ways that are small and are not particularly apparent in the view of the
plane presented. Therefore, in filling such a void through a small
passageway the mercury will fill the system indicating a pore volume
having the radius of the passageway. The mercury penetration data,
including this effect, are given in Figures 11, 12, 13, and 14. All of
the data have been normalized to 200 psi or 14 kg/cm2 . Three different

34 A.
A. Shidlovskii, "Principtee of Pyrotechnios", Air Force Systems
Comnand, Report No. AD-AO01859, 23 Oct 1974, Wright Patterson AFBJ OR,
Translation of .zd Masimoetroyeniye, Oxnovy Pirotekhniki, 3rd Ed.,
Moscow (1964).

i:1 38

JI
 ... ......

,4J

0Li
i

4-a

I'4-" V)
.9-

00

I ...
+C
U

EL

N co

I,)

01. 2O --

r=
C

_
.N

31VU
0c
V3HdS3W0.
00

r,, ..

al-Mmm 96mm
 whole grain samples of lots 6 and 10 were evaluated, Figures 11 and 12,
to determine the reproducibility of the sampling and measuring technique.
The penetration data for these samples shows fair agreement and most of
the mercury entered through passageways between 0.1 and 1.0 micron r'
meter. The experiment was expanded to include lots I and 11 and the..
data are presented in Figure 13 for whole grains. The experiment was
repeated on new samples of these lots where the grains were first frac-
tured in two. The penetration volume is presented for half grains in
Figure 14. The data for lot 10 appears to be in error, and for this
sample some of the grains may have been crushed by the applied pressure.
The mercury intrusion data for the four lots of black powder are
nearly the same for either whole or cut grains; neither group) showing
major differences in penetration pattern. On this basis, it is concluded
that the graph-Ite coating does not seal the grain to pressurized mercury.
The two experiments also show that mercury must be able to reach the in-
terior of a grain since cut and whole grains host the same v4,lume of
pene'ntration. If mercury intrusion had been related to the black powder
surface, then penetration between whole and cut grains would have been
approximately proportional to the ratio of the surface area of a sphere to
two corresponding hemispheres or 2/3. This was not found to be so and
the conclusion is that the mercury penetrated throughout the grain.I
The distribution of pass-Ageways was calculated using equation (2)

I and the data of Figures 13 and 14. The calculation is primarily depen-
dent on the slopes shown; therefore, it is not affected by the normali-
zaltion. The distribution is given in Figures 17 and 18. A correlation
exists between the number density of passageways between 0.2 and 0.1
micron and flame spread rate. No such relationship was evident for the
smallest of radii and this range will be discussed later.
D. B.E.T. Surface Area Measurements
The intrusion data can be used to calculate internal surface area
but because of the "link bottle effect" the result would be suspect. For
this reason the surface area was measured by the B.E.T. technique and
the results are given in Table 3 and plotted as a function of burning
rate in Figure 19. The standard deviation for four samples of lot 6 is
6.6, percent and similar agreement on replicates of the same sample was
found. It is clear that an increase of internal surface area increases
buraing rate. The graph, Figure 19, contains several markedly different
black powder samples including the jet-milled oak deviant lots, maple
GOEX Inc., powder and one CIL sample. The latter two producers used the
ball-wheel mill process. Overall, the samples appear to follow the same
. . function even in the small neighborhood tested where the density of black
powder does not vary to any great extent. it remains to the future to
develop a broader neighborhood for :nvestigation. Under these constraints
the relationship between burning rate and the degree of openness of a
grain of black powder seems reasonable.

40
 CLC

uii
¶i a
nn

a Cc

I-A

CUL

II 41
 p-C

C4

IA

- a U

(8) NlDn Nlnii 0

LU~ &ALU

_ 41.

MINE
 41..

4.1.

+ ~NO 1-

U,

10
+- 9 414*

I-+-

;+ wl 41

o 43
 The surface area of charcoal 2 2 is about 1000 22 gm- 1 and since
charcoal consititues 15.6 percent of black powder, an upper value of 156
m gm-I would be expected. The measured values are much less, about
0.5 to 1.0 m2 gm" 1 . Therefore, it is concluded that almost all of the
charcoal particles are sufficiently sealed to greatly reduce gas absorp-
tion. This could explain why burning rate was not found to be a function
22
of the original surface area of charcoal or other carbon materials.

It also should be pointed out that the surface area measurements

were extended from the original four samples to include all of the deviant
lots. This was performed as a matter of convenience rather than prefer-
ence over the pore size measurements, for at the time of analysis it was
more expedient to extend the surface area measurements.

E. Density Measurements

'rhe total free volume can also be calculated from the bulk density
of black powder, (IB p,), the density of the spparate ingi .dients, (DI),
aind percent compositIon. Such values were taken from Table 1 and the
following relationship, equation 3, was applied to a 100 gm sample. Ash
and moisture content were not included.

100100~KNOI+IS
++free volume, (3)
1)BP. 1)KNO DS DCJ

Using a density of 1.17 for oak and 1.43 for maple charcoal (determined
from charcoal extracted from lot 11) the total free volume was calculated.
Values are given in Table 4 and Figure 20. They incorporate both isola-
ted voids and accessible volume elements. Equation (3) can be used to
show that the density of charcoal is a sensitive parameter that affects
the free volume of black powder when a particular compression pressure
is selected for manufacture. Moreover, compressing black powder to a
particular density using different density charcoals will produce grains
with varying degrees of free volume. One problem yet to be addressed is
that no knowledge exists on the density variations for lot-to-lot pro-
curement of charcoal. If a variance is found, compaction pressure and
grain density will have to be adjusted accordingly to obtain the same
free volume or a particular density.

In an attempt to measure the volume accessible from outside of the

grain, the true density (DT) was measured by an ether displacement tech-
nique. These values and the bulk density (DB.p.) from Table 1, were
related to the free volume for a 100 gm sample by the equation:

100 -
100 free volume& (4)
BP T44
44
 4IA

tFM to 0

$.00

-j c.

r2. 000e~
-L

~I-I 45
 IA
090

0-

46J
 I1
These values are listed in Table 4 and in Figure 20. Comparison of the
free volumes obtained from equation (3) and (4) shows that in all cases,
except lot 4 and lot 11, the total free volume including voids is larger
than that accessible from outside of the grain. These volumes can be
compared to the mercury intrusion values. One comparison is to relate
the volume of' mercury penetrating through passageways larger than 0.1
micron, a plateau in Figure 13, to the density derived values. In this
case, the mercury values are smaller and this is reasonable as small dia-
meter pores have not been penetrated. However, examining the smallest
radii of 0.0)03 microns, the volumes are too large for lots I and 6. It
is therefore concluded that the sharp rising asymptote occuring at the
highest mercury pressures either elastically and/or inelastically con-
stricts the black powder grain. For this reas~n, the mercury intrusion
Sdat a i s suspec t beyond l 10,000 psi or 70 kg crm-

lrom the density measurements, both estimates of free volume are

.1 showtn as a tfuction of flame spread rate in Figure 20. The better cor-
relation is given by the ether displacement volume reflecting the volume
accessible from outside the grain.

If the volume occupied by moisture is considered, then it is clear

that a small percentage of water could be significant in relation to the
measured free volume of black powder. This could bg the physical effect
of water on burning rate. Shulman- 5 and Plessinger have measured the
effect of water on burning rate and their data Is reproduced in Figure
21. Since there could be a difference in equipment, these data sets
should be examined independently, but in either case the trend is clear
and it is suggested here, based on inference alone, that water plugs and
occupies passageways to retard burning rate.

V. FUTURE PLANS

,The suggestion is mtade that for future work a density range and
volatile content for charcoal be established. Also, compaction pressure
for making blacK powder shcald be selected at.' it is proposed that surfac(e
area and porosity measurements guide this study. In selecting compaction
pre•ssure, dic dimensions should be considered and the density distribution
determined, rhe density of biack powder is an indirect nea,cure of gross
structure being sensitive to the density of charcoal and particle size
of ingrdient•. From present work, it is proposed that the internal
structure be measured and controlled. In such a study it i.s proposed
to generate a standard meal and form black powder samples that providL
a greater variation in density, porosity and surface area to better
Sdocument the effect of th',se viriables.

*1 i

V. 47

. :%Un\-
, 4,

0 0U

", ~.1

4.
W ..,
CI

a,,

+ 00

i.F- 4 "0 '

 VI. SUMMARY

1. It was found that compaction of black powder meal produces con-

siderable plastic flow and forms a conglomerate matrix of interconnecting
passageways to the extent that the internal free volume is but a few
percent.

orrelations between internal surface area, pore volume, and

C2.
density were made with burning rate.

3. It was suggested that the adverse effect of water is the result

of occupying the free volume of black powder and retarding burning rate.

4. From the above relationships and S. P.M. nicrophotographs the

hypothesis was established that an increase in the degree of openness of
black powder grains increases burning rate.

AC KNOW LEDGEMEINTS

Assistance from several people was required to complete this study.

The hel1p and cooperation received was gratifying. I also drew extei-
sively upon Kevin White's (ABB) experience with black powder; his coinsel
and advice are a hallmark of this study. Other credit is given in the
text where appropriate; however I would like to mention here those
individuals who assisted in this work: Ralph Benck (PMIB), Dominic
Di Berardo (PNIB), John Gavel (Dupont), Gould Gibbons, Jr. (LEB), tileanore
Kayser (NSW), and Jean Owens (Micromeritics). Perspective was gained by
conversations with Hugh Kerr (GOEX, Inc.) and tHugh Fitler (ICIA) and I
thank both men for allowing me to tour their facilities. I also express
my appreciation to Eli Freedman (ABB) for his translation of part of a
French reference which is given as an appendix. I also thank Lois Smith
for helping prepare this manuscript.

4/S

- 7~J -.
 REFERENCES

I. Tadeusz Urbanski, Chemistry and Technology of IExplosives, Vol 3,

Pergamon Press, NY, pp. 322-346, (1967).

2. Arthur Marshall, Explosives, Sec. lHd. Vol 1, History and Manufacture,

P. Blakiston's Son and Co., Philadelphia, (1917).

3. B. T'. Fedoroff and 0. 1. Sheffield, incyclopedia of Explosivcs and

Re'lated Items, Vol 2., PATR 2700, Piic-inny Arscna'l, Dover, NJ,
p. B'IbS-B178 (1962).

4. Arthur P. Vangelder and Hugo Schatter, ilistory of the Explosive

Industry in America, NY Columbia University Press, NY (1927).

q. Chromalloy Corporation, "A Study of Modernized Techniques for the

Manufacture of Black Powder", Propcllex Chemical Division, Final
Report No. L)AI-23-072-5Ol-ORD-P-43, Jan 60, Chromalloy Corp., IL.

0. II. E. Carlton, B. B. Bohrer, and II. Nack, "Battelle Memorial Insti-

tute Final Report on Advisory Services on Conceptual Design and
Development of New and Improved Processes for the Manufacture of
Black Powder", Olin Corporation, Indiana Army Ammunition Plant, 20
Oct 1970, Charlesiown, IN.

7. J. R. Plessinger and L. W. Braniff, "Final Report on Development of

Improved Process for the Maunfacture of Black Powder", RCS AMURE-
109, Olin Corporation, Indiana Army Ammunition Plant, 31 Dec 1971,
Charlestown, IN.

8. Indiana Ammunition Plant, "Black Powder Manufacturing Facility",

Vol. I and 2, Indiana Army Ammunition Plant, 1975, Charlestown, IN.

9. Kjell Lovold, U.S. Patent No. 3660546, "Process for the Preparation
of Black Powder", 2 May 1972.

10. J. Craig Allen, "The Adequacy of Military Specification MIL-P-223B

to Assure Functionally Reliable Black Powder", Report No. ASRSD-QA-
A-P-bO, Ammunition Systems Reliability and Safety Div, Product
Assurance Directorate, June 1974, Picatinny Arsenal, Dover, NJ.

11. II. William Voigt, Jr., Pell W. Lawrence and Jean P. Picard, "Process
for Preparing Modified Black Powder Pellets", U.S. Patent, 3937771,
Feb 10, 1976, see also II. W. Voigt and D. S. Downs, "Simplified Pro-
cessing of Black Powder and Pyrotechnical Igniter Pills Including
Low Residue Versions", Seventeenth International Pyrotecinics Seminar
Vol. 2, ITT Research Inst., Chicago, IL, July 1980.

12. J. C. Allen, "Concept Scope of Work For MM4TE Project 57b4303

Acceptance of Continuously Produced Black Powder", Report No. SARPA-
QA-X-10, November 1975, Picatinny Arsenal Dover, NJ.

51

[. -,
"A • • .2•T •••- ... . " _ - .
 REFERENCES (Continued)

13. K. J. White, H. U. Holmes, J. R. Kelso Jr., A. W. Ilorst, and I. W.

May, "Ignition Train and Laboratory Testing of Black Powder",
ARRADCOM-BRL Report No. IMR-b42, April 1979, APG, MD. See also
K. J. White, H. H. lolmes, and J. R. Kelso, "Effect of Black Powder
Combustion on High and Low Pressure Igniter Systems", 16th JANNAF
Combustion Meeting, CPTA Publication 308, Dec 1979.

14. L. Shulman, private communication, (data quoted in reference 13),

15 Jan 1978, Picatinny Arsenal, Dover, NJ,

15. Hugh Fitler, private communication - Draft report, "Accuptance of

Continuously Produced Black Powder", 8 March 1979, ICI America Corp.,
Charlestown, IN.

16. Neale A. Messina, Larry S. Ingram and Martin Summerfield, "Black

Powder Quality Assurance Flame Spread Tester", Dec 1978, Princeton
Combustion Laboratories, Inc., Princeton, NJ.

17. R. A. Noble and R. Abel, Phil. Trans. Roy. Soc., London, Series A
Vol. 165, 49-155, (1875).

18. R. A. Noble and R. Abel, Phil. Trans. Roy. Soc., London, Series A
203-279 (1880).

19. F. A. Williams, "The Role of Black Powder in Propelling Charges",

Picatinny Arsenal Tech. Report No. 4770, May 1975, Picatinny
Arsenal, Dover, NJ.

20. James E. Rose, "Investigation on Black Powder and Charcoal", IHTR-

433, Sept. 1975, Naval Ordnance Station, Indian Head, MD.

21. James E. Rose, "Black Powder - A Modern Commentary", Proc. of the

10th Symposium on Explosives and Pyrotechnics, 14-16, 1979, Franklin
Research Inst. Philadelphia, Pa.

22. Abraham 0. Kirshenbaum, Thermochimica, 18, 113, (1977).

23. J. D. Blackwood and F. P. Bowden, Proc. Roy. Soc., London, A213,

285, (1952).
"4. Woldemar Hintze, Explosivstoffe, 2, 41, (1968).

25. L. Shulman, C. Lenchtiz, L. Bottei, R. Young and P. Casiano, "An

Analysis of the Role of Moisture, Grain Size, and Surface Glaze on
the Combusiton and Functioning of Black Powder", Picatinny Arsenal
Report No. 75-FR-G-B-15, Oct. 1975, Picatinny Arsenal, Dover, NJ.

26. L. Shulman, R. Young, E. Hayes and C. Lenchitz, "The Ignition

52
 REFERENCES (Continued)

Characteristics of Black Powder", Picatinny Arsenal Technical

Report No. 1805, Sept, 1967, Picatinny Arsenal, Dover, NJ.

27. Clyde Orr Jr., Powder Technol., 3, 117 (1969/1970).

28. Arthur W. Auamson, Physical Chemistry of Surfaces, 2nd lid., Inter-

science, NY, pp. 54o-549 (1967).

29. Ui.Alison Flood, Ed., The Solid-Gas Interface, Vol. 1, Marcel Dekker
Inc., NY, (1907).

30. R. NI. cilntosh and A. P. Stuart, Can. J. Res.., 1324, 124 (1946).

Il. Memorial Des Poudres lit Salpetres, Vol. 6, Chapter UI by M. Vieille,

Gatuthier-Vallars lit Fils, Imprimeurs-Libraires, Paris, pp. 256-391,
(189•3).

32. Naval Ordnance Laboratory, Ordnance Explosive Train Designers Hland-

book, NOLR 1111, April 1952, Naval Ordnance Laboratory, White Oak,

.I33. Rocbrt C. Weast, ed., Handbook of Chemistry and Physics, Slst lid.,
The Chemical Rubber Co., Cleveland, Ohio, (1970-1971).

3.v. A. A. Shidlovskii, "Principles of Pyrotechnics", Air Force Systems

Command, Report No. AD-AO01859, 23 Oct 1974, Wright Patterson AFB,
o0H, Translation of Izd Vo Masimostroyeniye, Osnovy Pirotekhniki,
5rd Ed., Moscow (1964).

5/

~J
 APPENDIX
PARTIAL TRANSLATION OF VIELLE ARTICLE-REFERENCE 30 PAGES 306-325
by Eli Freedman
4 TI. First Series--Agglomeration of Pulverized Materials

We have used the materials of RWP brown powder. Those materials,

ground and sieved through a silk gauze, were agglomerated dry by compres-
sion with a hydraolic press, using the steel mold already described.

We thus obtained pastilles of 20-mm diamet.-r, about 3.5 mm high; and

blocks of triple weight, 10.5 mm high. Six pastilles or 2 blocks of the
same density constituted a charge of loading density 0.6; 3 pastilles or
I block permitted the corrying out of comparative experiments at a loading
density of 0.3.

The sect ons of the pistons used for the measurement of pressure
were configured in a way to furnish similar crushings so as to make the
tracings comparable: a piston of 0.5 cm2 for the density 0.6, and the
2
olle of 1.33 cm for the density of 0.3.

The crushings were carried to 3.5 mm and 4 mm by the use of copper

cylinders similar to the regulation cylinders, but with a partially-
reduced crajss section, ratio of similitude /1V71.
Table 11, gives the results of the tests (only one half of the
table is reproduced here given as illustration).*

TABLE 11
Duration of Combustion
Density of Density of
Density Thickness Loading 0.6 Loading 0.3

1.922 10.44 98.07 125.0

1.785 11.19 87.13 108.80
1.7b0 11.47 48.33 74.57
1.724 11.65 17.36 28.78
1.635 12.09 5.22 12.01

The results o5 these 3 determinations are represented in Figure 15

(of text), whose abscissas represent the real densities of the pastilles,
and the ordinates, the durations of combustion. [Curve reproduced iii
Sassc's text as Figure 2].*

. rackets enctose tranesator'e commente cr notea.

4
P
J
A
/
 These four curves have a common shape. At low densities up to a
value of about 1.720, the duration of combustion is effectively inde-
pendent of thickness whether the combustion occurs at a charge density
of 0.3 under a maximum pressure of 1200 kg, or at a charge density of
0.6 under a pressure reaching 3000 kg. However, combustion time slowly
increases up to four times the combustion time of elementary dust; then,
suddenly, between densities of 1.720 and 1.800, the influence of thickness
suddenly manifests itself and the combustion duration increases with
extreme rapidity, tending to the limits characteristic of the compact
materials; i.e., effectively proportional to the thicknesses.

This effect of compression is hardly unique to brown powders. The

same functioning is observed with black powder materials reduced to the
powdered state.

V. General Mode of Action of Compression

In summary, for all of the materials that we have reviewed here,

the influence of compression on the duration of combustion occurs in an
identical way.

Four regimes of compressinn can bc distinguished which correspond

to four very different combustion modes.

In the first regime, with the weakest compression, the combustion

duration of the materials does not differ from that of the elements which
compose it placed next to each other. It is consequently independent of
thickness.

In the second compression regime, the combustion duration pro-

gressively rises up to values reaching 4 or 5 times those that corre-
spond to the grains of elementary dust, but this duration nevertheless
remains independent of the thickness. It ia thie type of materiaZ that
normaZly conetitutee the bZack or brown powdere actuaZZy used by artiZZery.

The third regime is characterized by an extraordinarily rapid

variation of combustion duration with density; the influence of thickness
appears and is accentuated more and more. This type of w.aterial is
sometimes introduced during manufacture, when compression is required to
achieve excessive slowing-up of the duration of combustion of the
elementary grain; but the extreme irregularity of the products does not
normally permit these types of materials to be used.

In the fourth regime, the materials, now virtually compact, show

durations of combustion slowly increasing with density and proportional
to their thicknesses.

, This succession of phenomena can be very simply explained, based on

the known laws of gas flows through capillary orifices. It is known that

451
the discharge, Q, of fluids across capillary canals of all kinds is given
by expressions of the type of Poisuille's law,

Q = KPI)k/I.

where I) -- diameter; P, pressure; L, length; and K, a coefficent charac-

teristic of the ýluid; and that consequently, the flow velocities, pro-
portional to Q/I)-, decrease with extreme rapidity with the dimension of
the orifices.

it is conceivable that in the phenomenon of the penetration of

incandescent [flamingJ gas across a heterogeneous mass, the interstices
of a certain order of magnitude, which will be called the first order,
alone play an effective role; the orifices of lower order, e.g., 10 times
less, permit the flame to propagate only with a velocity 100 times less.

In relation to the above ideas consider the material formed by agglo-

meration of grains at the lowest degree of compression. A system of inter-
stices exists in the charge which assures flame spreading to all of the
grains without a noticeable delay, or at least for a time duration negli-
gible with respect to the combustion time of the grains themselves. This
mode of combustion is made evident in the numerous examples given above,
showing the identity of the durations of combustion of the free grain
with the weakly-compressed blocks. This is the first phase of compression
at work.

Under stronger compression the interstices are reduced by a sort of

tangling of the grains, whose surfaces in contact become married in a
more and n.,)re complete way while blocking or reducing to a very small
dimension the entire grid of primitive interstices.
There results a new system of canals, whose dimensions remain of
the original order; rather thin limiting on the average each elementary
grain, as in the primitive state, they limit a polyhedral kernel formed
by the partial agglomeration of several grains; and it is readily con-
ceivable that the average duration of combustion of these kernels can
vary continuously with compression up to values double, triple, etc.,
the duration of combustion of the prir ,e grains without, however, a
residual system of first-order canals Leasing to exist in the mass, thus
assuring a combustion duration independent of the external dimension of
the grain. This is the working of the second phase of compression.

Under increasing compression, the number of first-order canals is

reduced enough so that the dimension of the kernels which they limit
become of the order of a milled grain. One falls therefore into the
irregularities which we have observed in the third phase of compression
of the raw materials forming the grains at the moment where the influence
of thickness is starting to be noticed. These irregularities continue

"57
until all of the first-order canals have disappeared and the influence
of thickness definitely shows up.
The fourth phase of compression starts at this moment. The materials
approach the limiting state corresponding to a perfect continuum and the
velocities of combustion vary only very slowly with density. It is in
fact conceivable that this velocity results from the conductibility of
the materials for the heat accruing from a slowa vetoo'tty of infiltration
of glow~ing gao across the second-order interstices, whose dimensions are
progressively reduced. From there on, there is a slow decrease of com-
bustion velocity, while the density rises up to the practically realizible
limits.
VI. Influence of Pressure on the Duration of Combustion
The results in Tables XI, XII, and XIII relating to the influence
of compression on the mode of combustion of the pulverized raw materials
permit the deduction of several qualitative conclusions about the influ-
ence that pressure exerts on combustion velocity.
If one considers materials of density greater than 1.780, for which
the proportionality of the duration of combustion to thickness is effec-
tively realized, i.e., the materials that function as co at one ob-
tains the following numbers [in the table] for the rat te durations
of combustion of the same elements to the charge densities 0.3 and 0.6
(maximum pressures 3200 and 1300 kg).
The general average of these numbers is about 1.25, and the exponent
of the pressure that takes account of this variation of the duration of
combustion is about 0.25. This value was previously noted in Chap. 11,
as resulting from the comparison of the durations of combustion of com-
pacted materials of 30/40 powder burning at densities 0.3 and 0.6.
When one considers materials of density less than 1.780, the ratio
rises rapidly and attains values greater than 2.00. Whatever the errors
which lead to irregularities in the ratios obtained in some of the pairs
of experiments, we can certainly conclude that the influence of presstmre
increases significantly when the compactness of the materials diminishes.
The ratio 2.5 effectively recapitulate~s the proportionality of the
durations of combustion to pressures.
The discrepancy of the pressure exponents given by experimenters to
represent the influence of pressure on the combustion velocity is easily
explained by the variable nature of the raw materials on which they
worked.
In any case, it is evident that this effect of pressure cannot be,
a prior'i, regarded as identical for powders of different factories, and
that it is important to determine it for each particular case.
(----page 325----)

58
 -7. t r- .a~ ...
-......- t-

* DISTRIBUTION LIST

No. of No. of
copies Organization Copies Organization

12 Commander 1 Director
Defense Technical Info Center US Army Armament Research
ATTN: DIDC-DDA and Development Command
Cameron Station Product Assurance Division
Alexandria, VA 22314 ATTN: DRDAR-GAS, C.J. Allen
Dover, NJ 07801
Director
Defense Advanced Research 6 Commander
Projects Agency US Army Armament Materiel
ATTN: LTC C. Buck Readiness Command
1400 Wilson Boulevard ATrN: DRSAR-LEP-L, Tech Lib
Arlington, VA 22209 I)RSAR-IRC, G.II. Cowan
DRSAR-LEM, R. Freedman
2 Director DRSAR-PBB-P, J. Hill
Institute for Defense Analyses DRSAR-PDB-P, D. Howard
ATTN: II. Wolfhard DRSAR-QAM-A, A. Bruskas
R.T. Oliver Rock Island, IL 61299
400 Army-Navy Drive
Arlington, Va 22202 1 Commander
US Army Watervliet Arsenal
Commander ATTN: Code SARWV-RD, R. Thierry
US Army Materiel Development Watervliet, NY 12189
and Readiness Command
ATTN: DRCDMD-ST 1 Commander
5001 Eisenhower Avenue US Army Aviation Research and
Alexandria, VA 22333 Development Command
ATTN: DRSAV-E
2 Commander P.O. Box 209
US Army Armament Research 1 St. Louis, MO 63166
Development Command
ATTN: I)RI)AR-TSS 1 Director
Dover, NJ 07801 US Army Air Mobility Research
and Development Laboratory
8 Commander Ames Research Center
US Arm>' Armament Research and Moffett Field, CA 94035
Development Command
ArTN: DRDAR-LCA, J. Lannon I Commander
DRDAR-LC, J.P. Picard US Army Communications Rsch
DRDAR-SCA, L. Stiefel and Development Command
DRI)AR-LCE, R.F. Walker ATTN: DRDCO-PPA-SA
DRDAR-LCA-G, D. Downs Ft. Monmouth, NJ 07703
DRI)AR-LCA-G, W. Voight
DRDAR-LCA-G, R. Field
DRDAR-LCA-G, P. Ostrowski
[Dover, NJ 07801

S9

I-
 DISTRIBUTrION LIST
No. of No. Of
Copies Organization Copies Organizatior
I Commander 1 Commander
US Army Electronics Research 4 US Army Research Office
Development Command ATTN: Tech Lib
Technical Support Activity P.O. Box 12211
ATTN: DELSD-L Research Triangle Park, NC 27706
Ft. Monmouth, N.1 07703
1 Director
1 Commander US Army TRAIOC Systems
US Army Missile Command Analysis Activity
ATTN: DRSMI-R ATTN: ATAA-SL, Tech Lib
Redstone Arsenal, AL 35809 White Sands Missile Range,
NM 88002
1 Commander
US Army Missile Command 2 Office of Naval Research
ATTN: DRSMI-YDL ATTN: Code 473, G. Neece
Redstone Arsenal, AL 35809 800 N. Quincy Street
Arlington, VA 22217
S1Commander
"115 Army Natick Research and 1 Commander
Development Command Naval Sea Systems Command
ATTN: DRXRE, D. Sieling ATTN: J.W. Murrin (SEA-62R2)
Natick, MA 01762 National Center
Building 2, Room 6E08
1 Commander Washington, DC 20362
US Army Tank Automotive Rsch
and Development Command 1 Commander
ATTN: DRDTA-UL Naval Surface Weapons Center j
Warren, MI 48090 AITN: Library Br, DX-21
J. East
1 Commander Dahlgren, VA 22448
US Army White Sands Missile
Range 2 Commander
ATTN: STEWS-VT Naval Surface Weapons Center
WSMR, NM 88002 ATTN: S.J. Jacobs/Code 240
Code 730
1 Commander Silver Spring, MD 20910
US Army Materials • Mechanics
"Research Center 1 Commander
ATTN: DRXMR-ATL Naval Underwater Systens Center
ýIatertown, MA 02172 Energy Conservation Department
ATTN: R.S. Lazar/Code 5B331
Newport, RI 02840

60
 DISTRIBUTION LIST

No. of No. of
Copies Organization Copies Organization

2 Commander 1 ARO Incorporated

Naval Weapons Center ATTN: N. Dougherty
A'I*"N: R. Derr Arnold AFS, TN 37389
C The lLen
China Lake, CA 93555 1 Atlantic Research Corporation
ATTN: M.K. King
ICommander 5390 Cherokee Avenue
Naval Research Laboratory Alexandria, VA 22314
ATI'N: Code 0180
Washington, PC 20375 1 AVCO Corporation
AVCO Everett Research Lab Div
Stiuperintendcnt ATTN: 1).Stickler
Naval Postgraduate School 2385 Revere Beach Parkway
ATrN: rech lib Everett, MA 02149
I). Netzer
A. Fuhs
Monterey, CA 93940
2 Calspan Corporation
ATTN: E.B. Fisher
A.P. Trippe
I
4 Commander P.O. Box 400
Naval Ordnance Station Buffalo, NY 14221
ATTN: C. Dale
C. Smith 1 Foster Miller Associates, Inc.
J. Rose ATTN: A.J. Erickson
Tech Lib 135 Second Avenue
Indian Head, MD 20640 Waltham, MA 02154
4 AFATL-DLDL 1 General Electric Company
AT'TN: 0. Heiney Armament Department
.. Widden ATTN: M.J. Bulman
N. Junno Lakeside Avenue
LT. Loverro Burlington, VT 05402
Eglin AFB, FL 32542
1 General Electric Company
2 AFOSR Flight Propulsion Division
ATTN: L. Caveny ATTN: Tech Lib
B.T. Wolfson Cincinnati, OH 45215
B~oiling AFB, l)( 20332
1 GOEX Inc.
2 AFRPL (DYSC) ATTN: Hugh Kerr
ATTN: D. George Berlin Plant
J.N. Levine Moosic, PA 18507
Edwards AFB, CA 93523
1 GOFX Inc.
1 Aerojet Solid Propulsion Co. ATTN: David Levey
ATTN: P. Micheli Route 3, Box 423
Sacramento, CA 95811 Cleburne, TX 76031
61

ik'.
 DISTRIWUTION LIST
No. of
ornanizationZatio
ie physics International Company
2 Hercules Incorporated 2700 Merced Street94577
Alleghany Ballistic Lab Leandro, CA
ATTN: R. Miller p.S
Tech Lib Opower Systems, Inc.
Cumberland, MID 21501 Ar L.C. Elmore

1Iercuie~ Incorporated g1$ American Street

BIcchus Work San Carlos,,A 94070
I science Applic~tiol5s TIC
ArTN: M. Beckstead
Magna, UT 84044 R.B. Wdelman
ATTN:
3 ICI Americas Inc. 23146 Cumorah Crest 91364
Woodland Ililis, CA
Indiana Army Ammunition Plant
ATTN: R.Sz2 Rockwell international Corp.
C'. mvlsey Rocketdync Division
ATTN: C. Obert
J.H, vlanagan
Charlestown, IN 47111
6633 Canoga Avenue
ICI Americas bLC. Canoga P.rk, CA 91304
Central
( Engineer'ing Department
ATTN: W.C. Brill Carp,.
Wilmington, DE 19897 2 Rockwell International
Rocketdyrne Division
ATTN: W. Haymes
Micromeritics Corp. Tech Lib
ATTN" I.I.Rootare 76657
Springs Rd. McGregor, TX
5680 Goshen
Norcross, GA 30093 Saafety Contulting Engineers, Inc.
ATTN: C.J. Dahni
Olin Corporation 5240 Pearl Street
Badg~er rmy Ammunition Plant 60019
Rosemontf IL
ATTN: J. WIRamnarace
53913
Barab~oo
B Shock Hydrodynamics, Inc.
ATTN: W.H. Anderson
Olin Corporation 4710-16 Vineland Avenue
New Haven Plant 91602
North Hollywood, CA
ATTN: R.L, Cook
27 D.W, Riefler Thio, 1 CorporatiCfln
.
Tli Divici
Winchester Avenue
"27S Wasat-h Division
New Haven, CT 06 04
ATflN: 1. Peterson
Tuzh Lib
Paul Gough Associates, Inc. P.O. Box 524
ATTN: P.S. Gough Brigham City, U'7 84302
P.O. Box 1614
Portsmouth, NH 03601

62
rnci",

,i~i.
 !('.I m~ Rp~tn'
[..I Mly• n I i' 'l V M i.m full

t! DISTRIBUTION LIST

No. of No. of
CO.•1. tCoies
2Orani -atlon Organization
l' I Thiokol Co-rporation ICase. Western Reserve Univ.
E~lkton Division Division of Aerospace Sciences
A`I"
ATN: F.. Sutton ATTNz J. Rfen
i Elkton, MI) 21921 Cleveland, Oil 44135

3 'hokol Corporation 3 Georgia Institute of Tech

lluntsville Division School of Aerospace Engineering
A'I'N -,). Flanigan
R. CI ick

1Ihntsvile, AL.
Iel'ch Lib
35807
ATTN: B.T. Zinn
E. Price
W.C. Strahle
Atlanta, GA 30332

I TRW Systems Group 1. Institute of Gas Technology

AT'TN: II. Korman ATrN: 1). Gidaspow
One Space Park 3424 S. State Street
Redondo Beach, CA 90278 Chicago, IL 60616

2Un1ited Technologies I Johns Hlopkins University/APL

Chemical Systems Div Chemical Propulsion Informa-
A'rTN: R. lrown tion Agency
Tech Lib ATTN: 1'. Christian
I1.0. Box 358 Johns Hlopkins Road
Sunnyvale, CA 94086 Laurel, MU 20810

I Universal Propulsion Co. 1 Massachusetts Institute of

ATTN: H.J. McSpadden Technology
1800 W. Deer Valley Road Dept of Mechanical Englneering
Phoenix, AZ 85027 ATTN: 1'.Toong
Cambridge, MA 02139
I Battelle lemorial Institute
ATTN: 'L'1 LibI Pennsylvania State University
505 King Avenue Applied Research Lab
Colulnhus, Ol 43201 ATTN: G.M. Faeth
P.O. Box 30
i Brigham Young University State College, PA 16801
Dept of Chemical Eingi neering
ATTN: N1. Beckstead I Pennsylvania State University
Pirov,o, UT 84601 Dept of Mechanical Engineering
ATTN: K. Kuo
I California Institute of Tech University Park, PA 16801
204 Karmar Lab
'I Mail Stop 301-4b
A
ArmN: I.l.C. Culick
1201 E. California Street
Pasadena, CA 91125

63

'tI
 DISTRIBUTION LIST

No. of No. of
CoisOrganization Copies Organization
2 Princeton University 1 University of Minnesota
Forrestal Campus Dept of Mechanical Engineering
ATTN: N. Messina ATTN: E. Fletcher
Tech Lib Minneapolis, MN 55455
P.O. Box 710
Princeton, NJ 08540 2 University of Utah
Department of Chemical Engineering
2 Purdue University ATTN; A. Baer
School of Mechanical Engineering G. Flandro
ATTN: J. Osborn Salt Lake City, UT 84112
S.N.B. Murthy
'rsPc Chaffee Hall 1 North White Plains Senior
West Lafayette, IN 47906 High School
ATTN: R.A. Nelson
Rutgers State University 96 Washington Avenue
Dept of Mechanical and North White Plains, NY 10603
Aerospace Engineering
ATTN: S. Temkin Aberdeen Proving Ground
University Heights Campus
New Brunswick, NJ 08903 Dir, USAMSAA
ATTN: DRXSY-MP, H. Cohen
SRI International DRXSY-D
ATTN: Tech Lib
333 Ravenswood Avenue ATTN: DRSTE-TO-F
Menlo Park, CA 94025A D RSAES -
Dir, USACSLI
Stevens Institute of Technology Bldg E3516, EA
Davidson Library ATTN: DRDAR-CLB-PA
ATTN: R. McAlevy, III
Hoboken, NJ 07030

University of California,
San Diego
Ames Department
AT'rN: F. Williams
P.O. Box 109
La Jolla, CA 92037
of Illinois SUniversity
Dept of Aeronautical Engineering
ATTN: Ii.Krier
Transportation Bldg, Rm 105
Urbana, IL 61801

64

d
F 7 --

USER EVALUATION OF REPORT

Please take a few minutes to answer the questions below;, ter u

this sheet. fold as indicated, staple or tape closed, and place
in the mail. Your comments will provide us with information for
improving future reports.

1. BRL Report Number____________________

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

3. How, specifically, is the report being used? (Information

source, design data or procedure, management procedure, source of
ideas, etc.)___________________________

4. Has the information in this report led to any quantitative

savings as far as man-hours/contract dollars saved, operating costs
avoided, efficiencies achieved, etc.? If so, please elaborate.

5. General Comments (Indicate what you think should be changed to

make this report and future reports of this type more responsive
to your needs, more usable, improve readability, etc.)______

6. If you would like to be contacted by the personnel who prepared

this report to raise specific questions or discuss the topic,
please fill in the following information.

Telephone Number:_____________ ________

Organization Address: _ _ _ _ _ _ _ _ _ _ _ _ _ _