UNCLASSIFIED
AD NUMBER
AD879156
NEW LIMITATION CHANGE
TO
Approved for public release, distribution
unlimited
FROM
Distribution: Further dissemination only
as directed by Office of Naval Research,
Washington, DC 20360; Dec 1970 or higher
DoD authority.
AUTHORITY
ONR ltr, 13 Jan 1972
THIS PAGE IS UNCLASSIFIED
nil nix our i
f.
<ma ceC
ii »
X
HPC 70-121
.X'.v/-)
/ /
I /
W •
r©/WASHIMSTON y^i¥HiiI¥ ^
/ ^ / • /
MR/ABPA ASSOOATEaSS
/ / /
CONTROLLED ORIENTATION OF DISCONTINUOUS FIBERS IN COMPOSITES
BY
T. L, TOLBERT
This document is subject to special export controls and each transmitted «o foreicn
®ovcranonts or foreign nationals may b© mado only with prior approval of the Director
of Materiel Sciences, Office of Nava! Research.
Q»
O
CJ5
<fiS5
PROGRAM MANAGER
ROUf BUCHpAHL
r
OKBANTO RESEARCH
A StlSScOtASY or- feSOJ-eSAmO CO*C?A?5Y ‘ v
sea n. lammen soule?a&d . 'st.\cuis. ^sso m earn
IeISMw efixS '
■e-
Ct
§ *f
I IK 'Hill
‘Jj ? f 1 ’
fir* *5M! to?? - ® j* 3
«tJfi i -f. $ ^
B s 4
i;
tli
' f «J*r, T< X * £
HPC 70-121
CONTROLLED ORIENTATION OF DISCONTINUOUS FIBERS IN COMPOSITES
T. L. TOLBERT
DECEMBER 1970
MONSANTO/WASHINGTON UNIVERSITY ASSOCIATION
HIGH PERFORMANCE COMPOSITES PROGRAM
SPONSORED BY QNR AND ARPA
CONTRACT NO, N0001A-67-C-0218, ARPA ORDER 876
ROLF BUCHDAHL, PROGRAM MANAGER
MONSANI0 RESEARCH CORPORA II ON
800 NOR'SH LINDBERGH BOULEVARD
SI. LOUIS, MISSOURI G5166
FOREWORD
The research reported herein was conducted by the
staff of the Monsanto/Washington University Association
under the sponsorship of the Advanced Research Projects
Agency, Department of Defense, through a contract with
the Office of Naval Research, N00014-67-~C~0218 (formerly
N0Q014-66-C-0045), ARPA Order No. 876, ONR contract
authority NR 356-48/4-13-^6, entitled "Development of
High Performance Composites.."
The prime contractor is Monsanto Research Corporation.
The Program Manager is Dr, Rolf Buchdahl. (Phone: Area
Code 314-694-4721).
The contract is funded for $7,000,000 and expires
30 April 1972.
CONTROLLED ORIENTATION OK DISCONTINUOUS FIBERS IN COMPOSITES
T. L. Tolbert
ABSTRACT
Yarns of uni.directional.ly oriented, discontinuous high-
modulus fibers and whiskers have been successfully produced
in the laboratory by a modified vortex spinning process. Both
core yarns in which the higher modulus fibers are overwrapped
with organic or small diameter glass fibers and all-whisker-
yarns have been spun. These yarns can be readily incorporated
into plastics as reinforcing agents by filament winding and
related procedures. The strength of resulting composites is
markedly superior to those of similar materials fabricated by
other technique! due to well controlled fiber orientation and
very uniform fiber overlap.
fONTUOU.KD OK i KMTATION OK H I SC'ONT I NtlOUS K I UK US IN COMi’OS I TKII
T. !,. TOIjHEKT
Monsaiito/W.ishi r.gt on tin Ivors i fy Associal i.on
St. l.oui Missouri U.S.A.
The t x'nH'mloas potent i n 1 ol high modulus t i bri s and
whiskers as r»' i.iit.orci nq agents lor plant ics ,ind metals i.s now
well recognized. In a little inon' than a decade, use of such,
materials has advanced from laboratory studies aimed primarily
at extending glass fiber reinforced plastics technology, or
imitating naturally occurring fiber reinforced systems, to com¬
mercial manufacture and acceptance of resulting composites as
practical engineering materials. Composites containing high
modulus fibers such as boron and graphite are currently employed
as primary and secondary structure in such diverse applications
as components for military and commercial aircraft, sports
equipment and advanced prosthetic devices.
Key to the almost explosive growth of the composites field
has been development of effective methods for producing and
utilizing high modulus fibers. Small scale manufacture of con¬
tinuous fibers such as boron and graphite fiber was achieved
quite early and production of high quality fiber made routine.
More recently, procedures for much larger scale operation have
been dove loped which promise to reduce cost and make use of
composites reinloieod with t hose ? i be n; tenet.si tv pr.iet i ca I. ioi
)
many orw iiu'ivr iruj appi. ! •' ** i ana . Methods tor Labrica tiny boron
and qrnph.it>' t.' i in' composites, one;' primarily hand operations ,
have been greatly improved and to some extent automated and
reliable procedures worked out tor incorporating and composites
into component structure. Utilization of boron and graphite
composites thus now appears more limited by shortcomings in matrix
performance and in design methodology than by deficiencies of
the fibers.
Significant advances have also been made in employment of
whisker fibers as reinforcing agents, although overall progress
has not been as great as with boron and graphite. The. concept
of using monocrystalline fibers as reinforcing agents is extremely
attractive, of course, and so has been the subject of some ex¬
cellent research, especially in the U.K. Good progress has been
made in solving problems related to whisker fiber quality, in
scaling-up manufacture and even in bringing the price of such
materials to practical levels. For example, Norton Aesearcn
Corporation, Cambridge, Mass., North American Licensee of the
Phillips AG Silicon Carbide Whisker Process, is said to have
under development a process for producing extremely uniform
good quality whiskers on a pilot scale at projected commercial
prices of less than $50.00 per pound. Uni or tunate 1y, however,
progress toward utidi -’.ing whiskers as reinforcing agents has
teg** ifrteim i ‘Win n i ■
been much slower; pract ical, methods are still needed for con¬
verting whiskers and resin into high performance composites.
As single crystals, whiskers exhibit equivalent moduli
and far higher strengths than other fibers of similar composi¬
tion. It was the strength of these fibers, often more than a
million psi, that initially generated so much entnusiasm for
the idea of whisker reinforced systems. Exploiting this is
another matter, however; whisker strength has proved to be
extremely difficult to translate into composite strength. dnlike
boron filament and graphite fibers, whiskers currently are pro¬
duced only in discontinuous form and are usually extremely small
in size, often only a few microns in diameter and a few millimeters
long. Both factors greatly complicate the problem of controlling
directional orientation, placement, overlap and distrioution of
the fibers in composites to the degree necessary for good reinforce¬
ment efficiency, and hence good composite properties. Control
of fiber orientation has proved particularly difficult. As would
be expected, these problems are not limited to whiskers, but
also have been encountered in attempts to employ a variety of
other small diameters discontinuous fibers as reinforcing agents.
A number of approaches to overcoming these fabrication
problems have been reported, the two most effective being in-situ
growth of reinforcing whiskers in a metallic matrix and (low
T
- 4 -
orientation of short, high modulus fibers in a viscous, coagulabte¬
res in. The latter approach, pioneered with whisker ad a asbestos
reinforced systems by workers at the Explosives Research and
Development Establishment, Waltham Abbey, Exxex, England, has
proved particularly versatile. Strands of oriented fibers suitable
for subsequent processing can be prepared in this way by "wet
„ 1-4
spinning techniques or, at the other extreme, finished com¬
posites in which the fibers are flow oriented can be prepared
directly by modified injection molding procedures.^ Unfortunately,
these techniques are not suitable for processing high modulus
fibers greater in length than about 10 mm, and so development
of an alternate method for handling these materials has beer,
necessary. The technique found most successful in studies
conducted by the Monsanto/Washington University Association is
based on a textile process, "vortex" or "open end" spinning;
this is described below.
Spinning of Whisker Fibe r Yarns
Impetus for development of an alternate method for pro¬
cessing discontinuous high modulus fibers, such as whiskers,
into composites came in cur program from a study on the
reinforcement efficiency of such fibers. Long staple (10-10 nun)
3-silicon carbide whiskers had been made avai lable 1 or the work,
but no satisfactory method was available lot converting them
into the required uniditeet i i >n t! ly reinforced pi.isl ic specimens.
While flow orientation techniques were quite satisfactory for
preoaring directionally reinforced composites from short, fibers,
scuh an approach could not be employed with the longer material
due to excessive breakage of fibers caused by the high shear
forces generated in these systems. Following the lead of
6 1
several earlier workers ' in the U.S.A. , attention was given
instead to study of several textile processes which might be
modified to produce resin-free "packages" or yarns of axially
oriented fibers suited to filament winding and related composite
fabrication procedures. The most promising of these proved to
be a form of "vortex" or "open end" spinning.
Initially, a simple laboratory vortex spinner, similar in
principle to that developed by Strang for textile yarns, was
used to prepare yarns containing whisker fibers entrapped among
helically twisted textile (binder) fibers. In this, fibers dis¬
persed in a viscous medium were formed into a strand by the action
of hydrodynamic forces on the suspension as it passed through a
rapidly rotating tube in the spinner head. The yarn was collected
as it emerged from the tube assembly, washed free of the sus¬
pending liquid (usually corn syrup for convenience) in a gentle
stream of solvent, and taken up on a rotating drum. Yarns were
reasonably uniform and, even though the binder 1 Leers were un~
6
crimped, of sufficient strength tor composite preparation. Un¬
fortunately, however, axial orientation of the entrapped whisker
fibers was quite imperfect due to a tendency of the whiskers to
follow the helical path of the oinder material. In addition,
fiber breakage was excessive.
Modification of the first apparatus to permit introduction
(
of whiskers as a separate, more viscous suspension at tfia center
of the flowing binder fiber suspension prior to entry to the
rotating tube assembly proved a more satisfactory approach. Fluid
forces are not great enough in the central region of the sus¬
pension to disturb orientation or cause breakage of the whiskers
as they are introduced, compacted and wrapped by the binder
material. Resulting yarns, consisting of a core of axially
oriented whiskers helically wrapped by a thin layer of textile
fibers, are uniform and strong enough for easy handling. WhisK«?r
breakage is minimal.
The modified spinning apparatus is shown in Fig. 2. It
consists of a vertically mounted vortex spinning assembly led
under nitrogen pressure from spearato holding tanks for the
fiber slurries. The spinning assembly is basically; like that
in the previous apparatus except tii.it a short portion of the
spinning barrel i mined i a l e 1 y be low the i ot at leg lube is expanded
i n i u ni i u i in -nrrdttflt’Mtrfti
7
to approximately twice the original diameter to provide a chamber
for combining fiber slurries. In oar particular system, core 1
slurry is introduced through a 6 sum tube, just below the rotating
/
tube assembly, into the center of a 16 mm slurry chamber.
The slurry chamber and core slurry tube are connected directly
to the holding tanks by tubing entering through the top of the
tanks and ending in a conically flared tip about 12 inni above
the bottom. Lids of the tanks are fastened with bolts and are
sealed with rubber "Q" rings capable of withstanding 100 psi
pressure. In operation, the speed of the rotating tube and
the rate of slurry flow are controlled (annular flow must exceed
core flow) so that the fibers twist together in the short con¬
verging section of the chamber, rather than in the spinning
barrel proper, since in this region the outer fibers approach
tf*e core from a non-axial postion and are wrapped more uniformly
around the core.
The speed at which the rotating tube of the spinner head
must turn in order to impart enough twist for good yarn strength
varies with each system, depending primarily on the flow rates
of the slurries, slurry viscosities and the mechanical properties
of the fibers. With our apparatus, rotational rates ranging from
1 200- .1700 rpm produce sufficient twist in most cases for yarn
take-up velocities of several inches par second. Faster spinning
is cadi Iy possible but requires more precise control of variables.
H
Due to the high cost of whiskers, various glass Libers
were used as model materials in developing the. spinning ap¬
paratus and in all process studies. See Table 1. These
proved to be excellent substitutes for whiskers; no difficulties
were encountered in switching the spinning operation from one
material to the other. Several of the glass fiber yarns also
proved to be interesting in their own right. It was found, for
example, that the diameter of Beta glass fibers is small enough,
and fiber flexibility therefore great enough that these fibers
can be used both as wrapping and core materials. Thus, yarns
l
of short "E" glass fibers wrapped with Beta glass fiber and of
100% Beta fiber can be prepared. The textile fibers employed
as binder materials included uncrimped acetate, triacetate,
high tenacity rayon and Acrilan 57. Ail were found to be
satisfactory for the purpose.
In producing core yarns, approximately equal concentrations
of glass or whisker core fiber and of the lower modulus binder
fibers were used in the spinning slurries. Maximum concentration
was determined by the amount of the binder fibers which could
be used since these are the more flexible and exhibit tne greater
tendency to entangle with resultant Loss of dispersion uniformity.
For the lour binder materials shown in Table i, the limiting
9
r
I
i
i
concentration for 18 nun staple dispersed in corn syrup (100-350
poise) ranged from 0.05 to 0.1 volume percent when the spinning
tube of the Laboratory equipment was 6 mm in diameter. In order
to minimize damage to the fibers, the viscosity of the core
slurry was maintained at approximately 1.00 times that of the an¬
nular slurry.
Concentrations of high modulus fibers in the vortex-spun
yams prepared in our studies ranged up to about 36 voluraer
percent, the degree of axial orientation being very high at all
levels. As mentioned, with, smaller diameter glass fibers it
also was possible to spin 100% glass staple yarns of reasonably
good uniformity and strength. Similar, although much weaker,
100% yarns were spun from whisker fibers as well. The supply
of whiskers was not sufficient to permit, more than a few attempts
to prepare such yarns, but results were promising enough that
i
there is little doubt that conditions could be found for producing
good quality all-whisker yarns.
Summarizing efforts on whisker yarn preparation, a
process has been.worked out for converting discontinuous, high
modulus fibers into a yarn consisting of a core of these
fibers in axially alig> ed form helically wrapped with lower
modulus binder fibers. The yarn is produced by subjecting a
)
5
S 0
two component suspension, made up of a viscous, familiarly
flowing slurry of high modulus fibers urrounded by a more
rapidly flowing, less viscous suspension of much lower modulus
fibers, to an axial vortieity gradient. The gradient, generated
in the suspension by flow through a rotating tube, provides the
twisting force required. Core yarns containing up to 36 volume
percent of long staple 8-silicon carbide whiskers and even higher
percentages of glass fiber have been prepared in this way. Strong,
relatively uniform textile-like yarns of discontinuous Beta glass
fiber alone also have been produced in this way.
Vortex spinning of inorganic fiber core yarns is the only
potentially practical approach known to the author by which
long staple small diameter high modulus fibers and whiskers
can be '"packaged" in polymer-free directionally oriented form.
While strictly a laboratory technique as practiced here and of
doubtful interest for textile applications, this or a related
approach could be the key to much larger scale utilization of
such fibers as reinforcing agents. Certainly, it is one of the
first methods* to make filament winding and related fabrication
* The Canadian "Bobtex" Integrated Composite Spinning process' is
said to be applicable to some inorganic fibers and so, assuming
that these are fibers of interest as reinforcing agents, may be
another way of obtaining yarns for composite fabrication. However,
since a binder polymer must be presen' in "Bobtex" yarns to sup¬
plement the twist-applied forces between libers, use may be re¬
stricted in some systems to composites of relatively low fiber
con to lit .
11
procedures worth considering as means for controlling the
directional orientation, placement and spacing of whisker fibers
in composites.
Co mpo sites From Whi ske r Ya rns
Laboratory preparation of composites from thermosetting
resins, such as the lower viscosity epoxies, and vortex-spun
whisker fibers is straight forward using either compression
molding or simple filament winding procedures. Resin wet-out is
easy, so long as yarn twist is not excessive, due to ready sorption
and wicking of the uncured liquid resin by the binder fibers.
Bonding between the binder material and the matrix resin is good
enough under most conditions that these fibers actually behave as
if they were part of the matrix itself.
The orientation distribution of reinforcing fibers and
overall quality of composites prepared with vortex-spun yarn
are fully equivalent to those which can be achieved by very
carefully controlled flow molding. The orientation distribu
tion of glass fibers in a typical yarn reinforced molding is
shown in Table 2. Vortex, yarn composites are distinguished
from other systems, however, by the markedly improved fiber
overlap achieved. Overlap of the fibers in these yarns is
statistically uniform, while in extrusion and flow molding
12
processes fiber overlap is more random and depeiftis ent irely
on shear forces acting on fiber bundles and matrix. The result
is that, the yarn reinforced composites exhibit significantly
hiyhe: tensile strengths than flow molded materials, as shown
in Fj.g , 4for composites of epoxy resin and 3/8 inch glass
fiber (nominal length). In fact, with the exception of
specially prepared model specimens, the yarn reinforced
materials proved to be stronger than experimental composites of
equivalent composition made by any other technique in our
laboratory. Composite moduli are much less sensitive to localized
discontinuities and so are roughly equivalent for both yarn
reinforced and flow molded systems. The moduli of the experimental
composites from Fig. 4. thus can be plotted on a common line, as
shown in the comparison with "rule of mixture" values in Fig. 5.
Studies of whisker fiber composites prepared from vortex-
spun yarns are still in progress and data not yet complete
enough to justify publication. However, all indications to
date are that improvements in composite strength are similar to
those obtained in the glass systems. Reinforcement efficiencies
calculated as the percentage of fiber strength utilized in the
composite are much lower for the whisker reinforced systems,
of course, but absolute strength markedly improved.
13
Ac knowl edgmen t
A number of workers in the Monsanto/Washington University
Association have contributed to the work which has been
described. Particular credit is due to Dr. Myrne Riley for
initial development of the vortex spinning process for high
modulus fiber yarns and Dr. T. B. Lewis for the data on
composite properties. This work was performed by the Monsanto/
Washington University Association sponsored by the Advanced
Research Projects Agency under Office of Naval Research Contract
N00G14-67-C-Q218, formerly N00014-66-C-0045.
14
REFERENCES
1. Parratt, N. J. Composites, 1 [1], 25 (1969); 1 [31
141 (1970).
2. Hollingsworth, B. h. , Co mposi tes, 1 [2], (1969).
3- Schierdinq, R. G. and Deex, 0. D., J. Composite Materials,
3, 618 (1969) .
4. Schierding, R. G. and Tolbert, T. L., U. S. Government
Report, AD 866, 167.
5. Goettler, L. A., Modern Plasti cs, 47 [4], 140 (1970).
6. Garde11a, J. W. and Freeston, W. D. Jr., U. S. Government
Report, AD 825, 310L.
7. Shaver, R. G., U. S. Government Reports, AD 826. 346 and
AD 839, 519.
8. Strang, P. M., U. S. Patent 2, 700, 866; America's Textile
Reporter , Part I, January 2, 1964 and Part II, January 9, 1964.
9. Bobkowicz, E. and Bobkowicz, A. J,, Canadian Textile Journal,
April, 1970. Vol. 87 4, pp 77-84.
TABLE 1
FIBERS USED
IN YARN SPINNING
Fiber
Average
mm
Diameter
mils
E-glass
0.013
0.5
S--glass
0.0105
0.413
Beta glass (Owens-Corning
Fiberglas Corp.)
0.00414
0.163
Triacetate
0.0239
0.94
Acetate
0.0211
0.83
Acrilan 57A {Monsanto)
0.0183
0.720
Fortisan Rayon (Celanese)
0.00919
0.36
!
TABLE 2
ORI ENTATION OF GLASS FIBE RS IN A TYPICA L Y ARN COMPOS ITE
Molding Technique Hand. Lay-up/Compression
Reinforcing Fiber 0.163 mil E-glass
Matrix Shell's Epon 823 Z
es off Axis)
% Orientation
0-5'
6-10
11-15
16-20
21-25
26-30
31-35
36-40
41-45
46-50
51-55
56-60
61 - 6 5
66-70
71-75
76-80
81-85
12.5
10
12
11.6
10
9
7.7
3.3
4.2
2.4
3.9
2.2
3.4
1.5
2.9
1 . 7
(. . 7
1 . 0
86-9 0
SCHEMATIC OF OPEN
SCHEMATIC OF OPEN-END COMPOSITE SPINNING APPARATUS
VARIATION OF COMPOSITE STRENGTH WITH FIBER CONTENT
*
T 4 ” 25° C
"ter-an--1 d“ .to
VOLUME FRACTION (F FIBERS
S«*t Ilf 1 f V ( IjsMtu Oil _ _
"”” ""document control data r & i>
''■ciurifv < linix^N TiOfiin </ /»?/«*. /•<-»>> of ,<tt>-\ litu t .>« > m f Unit- « #«lj» .M/rin',iin.ii uiu'-f >*'< »/»rn T/i«- nmul.1 f**jn»s< I ■■ i f,i\.w(fr</)
* I'M-;.. AC IUM V (t'nilfitffMlr fllifftori .»*. ID I'OH I SI cum 1 V ( l ASSM U A IK»N
UNfl ASS IF HD
Monsanto Research Co: poration ;vr;To^‘ V ^ 1 —
Controlled Orientation of Discontinuous Fibers in Composites
Jt'SCMIf* llVt MOTES ( TVjK* ul rrpurf <«lif #rtr fu.l f v«* </«(***)
AU ThORiSl (First nmarr, mufdle initial, tamt name)
Thomas L. Tolbert, Monsanto Research Corporation
fi HEPOnr Date
De cembr > r 19 70
0 * 9 . CONTRACT OR GRANT f.O
NO0014-67-C-0218
l. PROJEC T no.
7j». TOTAL MO. OF f’AGES /6. MO OF REF5
31 9
Ua. ORIGINATOR’S REPORT N UM BF H|S)
HPC 70-121
Ob. OTHER REPORT NOISI (Any other number* that inny be assigned
this report)
ui outribution statement This document is subject to special export controls
and each transmittal to foreign governments or foreign nationals may
ae made only with prior approval of the Director of Material Sciences,
ffice of Naval Research, _ _ _
12 SPONSORING MILITARY ACTIVITY
Office tff Naval Research
Washington, D. C. 20360
V)Yarns of unidirectionally oriented, discontinuous high-modulus
ibers and whiskers have been successfully produced in the laboratory
y a modified vortex spinning process. Both core yarns in which
he higher modulus fibers are overwrapped with organic or small
iameter glass fibers and all-whisker-yarns have been spun. These
arns can be readily incorporated into plastics as reinforcing agents
y filament winding and related procedures. The strength of resulting
omposites is markedly superior to those of similar materials
abricated by other techniques due to well controlled fiber orientation
nd very uniform fiber overlap. . -
