Mechanical Testing Methods of Fibers For carbon and graphite yarns, strands, rovings,

and tows, the resin impregnated tow tensile test

and Composites (ASTM D4018) is recommended. The purpose of the
impregnating resin is to provide the tow with sufficient
Commonly employed techniques to determine the mechanical strength to produce a rigid test specimen
mechanical properties of fibers and fiber-reinforced capable of sustaining uniform loading of the individual
metal- and polymer-matrix composites (MMCs and filaments. Care should be taken to align the fibers
PMCs, respectively) are briefly described in this article, prior to impregnation. A resin compatible with the
some of which are also applicable to ceramic-matrix fibers and in minimum required quantity to produce
composites (CMCs). Wide coverage of these test the tensile specimen should be used. The failure strain
methods is given in ASTM standards (ASTM Stan- of the resin should be significantly higher than that of
dards 1998) and in sections of this encyclopedia on the fibers. End tabs (cast or mounted) may be
composite materials (see Composite Materials: OŠer- necessary in some instances for alignment purposes
Šiew). and to prevent slippage at high loads.
Creep testing of fibers is carried out using dead load
tests at constant temperatures (either in air or in a
suitable environment). A length of fiber is gripped
1. Fibers vertically in cold grips and a resistance furnace is used
The properties of interest at low temperatures are the to maintain constant temperature over a specified guage
elastic modulus and the fracture strength whereas for length (typically 25 mm) of the fiber. Elongation and
high temperature applications and long-term property fracture strain are measured using any of the non-
modeling, creep properties such as the creep–rupture contacting displacement devices. Bend stress relax-
strength, time- and strain-to-failure are necessary. ation is used for measuring the creep behavior,
Most fiber properties are probabilistic in nature, particularly for small diameter fibers. This method
hence, many fibers have to be tested and the data consists of tying the fiber into a loop and exposing it to
generated should be statistically analyzed. It is recom- the appropriate temperature for a specific time. After-
mended that the fiber properties be measured on wards, the diameter of the loop is measured before
a material that represents composite behavior. For breaking it. The residual loop radius is used to estimate
example, fibers extracted from as-processed or fa- the creep behavior.
tigued composite specimens should be tested to
obtain properties after processing or fatigue testing,
respectively.
The standard method for determining room tem- 2. Composites
perature properties is the single-filament tensile test 2.1 Tension
(ASTM D3379). In this method, a random selection of
fibers is made from the material to be tested. The fibers Straight-sided dog-bone type loaded specimens with
are centerline-mounted on special slotted tabs (Fig. 1). rectangular cross-section are commonly used for
The tabs are clamped so that the test specimen is measuring the in-plane tensile properties of unidirec-
aligned axially in the grips of a test machine and then tional, as well as cross-ply, composite laminates
loaded to failure at a constant displacement rate. It is (ASTM D3039 for PMCs and D3552 for MMCs).
important to measure the cross-sectional area care- Specimens can be end-tabbed if necessary to distribute
fully. The axial strain is measured using a noncontact- the load from the grips (either wedge or hydraulic) into
type extensometer. the specimen with a minimum stress concentration.
Test specimens with round or tubular cross-sections
can also be used. An important factor for obtaining
highly accurate test results is minimization of the
specimen misalignment with the load train and the
resultant bending stresses. Likewise, good alignment
between the desired fiber orientation and the specimen
axis is a must.
The following alternative methods are also avail-
able.
(i) The sandwich beam test (ASTM C393) for
evaluating the flexural properties of flat sandwich
constructions in which the test laminate is bonded to
either side of a suitable honeycomb core and the
sandwich is subsequently tested in a four-point flexure
Figure 1 configuration. To minimize thermal expansion mis-
Fiber testing method. match problems, the thickness of the laminate on the

1
Mechanical Testing Methods of Fibers and Composites

Test machine adapter

Bearing post

Lower grip

Upper grip
with
linear bearing

Specimen

Lower
grip
header Adustable jaws
tightenend by
thumbscrews

Specimen alignment pin

Baseplate

Figure 2
The Iosipescu shear test.

compression side should generally be twice that of 2.2 Compression

the tensile side. This test is also useful for tensile
Compressive testing of composites is relatively more
testing, particularly for the 90m properties of unidirec-
difficult than tensile testing. The mode of failure
tional composites.
involves some type of buckling, ranging from classical
(ii) Split-disc ring tensile test for apparent tensile
column buckling of the entire specimen to local micro
strength of ring or tubular composites (ASTM D2290)
buckling of the individual fibers. The test fixture should
in which a hoop-wound narrow ring is loaded using a
be appropriate to avoid buckling. Many axial com-
split-disc loading fixture that applies a hoop stress to
pression test methods in current use are some variation
the test ring.
of the Celanese compression test (ASTM D3410). In
The out-of-plane tensile properties of composites
this method, a thin, straight-sided specimen with a
can be evaluated by direct out-of-plane loading using
small guage length (" 13 mm) is loaded in the tabbed
a laminated specimen bonded between two fixture
region using inverted cone wedge grips. The IITRI
blocks (square or circular cross-sections) that are
method (ASTM D3410), which is a minor modifi-
pulled in tension. Alternatively, curved composite
cation, incorporates flat wedge surfaces rather than
beams, which take advantage of the out-of-plane
conical surfaces for greater stability. An untabbed,
tensile loading induced in the elbow of a curved
thick, straight-sided composite specimen can also be
laminate beam subjected to an opening moment, can
used, end-loaded in compression and side-supported
be used for indirect measurements.
to control lateral buckling (ASTM D695). The end-

2
 Mechanical Testing Methods of Fibers and Composites

loading might induce deformation and minor crushing

2.5 Fatigue
at the ends, however these are not of much concern if
strain guages or extensometers are used to monitor Fatigue in composites is application dependent, hence
strains rather than cross-head displacement. This the laminates should represent the application and
method has the advantage of requiring a simpler test testing should account for the loading spectra and
specimen and test fixture than the Celanese config- environment experienced during service. Unnotched
uration. The compression specimen geometry and fatigue tests on specimens similar to those used for
method of load introduction should be relevant to the tension tests can be conducted, either under load or
actual application for which the data is being gen- strain control and either triangular or sinusoidal
erated. waveforms (ASTM standard D3479 for PMC and
E466 for MMC). High loading frequency (
10 Hz)
may lead to frictional heating at the fiber\matrix
interface due to relative sliding. Fatigue crack growth
2.3 Shear tests are conducted using either center-notched or
The in-plane shear response of the composites can be single edge-notched specimens, preferably with fixed
obtained by tensile testing of a p45m laminate (ASTM far-field stress amplitude. Compact-tension (CT)
D3518). This test is simple and requires little fixturing. geometry can also be used if the composite laminates
The Iosipescu shear test (ASTM D5379) utilizes a V- have sufficient thickness. The data are typically pres-
notched beam wherein a material coupon in the form ented as crack extension versus the number of fatigue
of a rectangular flat strip with symmetrical centrally cycles. By use of suitable fiber bridging models, the
located V-notches is loaded using a special test fixture data can also be converted into crack growth rates
with the notches located along the line of action of against the effective stress intensity factor range.
loading (Fig. 2). Two strain guages, oriented at p45m to
the loading axis and located in the middle of the
specimens and along the loading axis, are used for 3. Interfacial Properties
measuring the shear response. By changing the ma-
terial coordinate system with respect to the loading Fiber\matrix interface properties such as the debond-
axis, both in-plane and out-of-plane shear properties ing strength and sliding resistance play a predominant
can be evaluated using this technique. Additional role in determining the properties of the composites.
methods such as the rail shear test (ASTM D4255-83), Fiber push-out testing is a commonly employed
the ten-degree off-axis shear test, and the tube-torsion technique for measuring these properties. In this test,
test (ASTM D5448) are also available. The out- an indenter (punch) is used to apply axial compressive
of-plane shear properties can be evaluated by using loading to a selected fiber in order to debond the fiber
the short-beam strength test (ASTM D2344) or the and force it to slide relative to the matrix. The
Iosipescu test. specimen, a thin slice of the composite (typically
0.6–1.3 mm thick) with the fibers oriented axially
within p1m, is mounted on a support plate that has
grooves to facilitate the fiber once it has been pushed
2.4 Fracture Toughness out from the opposite face of the sample. A detector
can be used to monitor acoustic emissions associated
The double cantilever beam (DCB) test (ASTM with the debonding events. Composites that have high
D5528) is the most widely used test method for interfacial strengths are not amenable to this tech-
measuring Mode I fracture toughness of unidirectional nique. For composites that exhibit single or multiple
composites. The specimens are " 125 mm long, 20– matrix cracking, alternative techniques include single
25 mm wide, and 3–5 mm thick. One side of the fiber pull-out and hysteresis measurements.
specimen is notched (" 25 mm deep) and the cracked
faces pulled apart with the aid of either piano hinges or
loading blocks attached to the specimen on the cracked
end. Load–displacement data are recorded and used for Bibliography
computation of the energy release rates. For Mode Annual Book of ASTM Standards 1998 Vol. 15.03, American
II toughness, the end notched flexure test, which uses a Society for Testing and Materials, Philadelphia, PA
specimen similar to that of the DCB test but loaded in
three-point bending configuration, can be used. U. Ramamurty

3
Mechanical Testing Methods of Fibers and Composites

Copyright ' 2001 Elsevier Science Ltd.

All rights reserved. No part of this publication may be reproduced, stored in any retrieval system or transmitted
in any form or by any means : electronic, electrostatic, magnetic tape, mechanical, photocopying, recording or
otherwise, without permission in writing from the publishers.
Encyclopedia of Materials : Science and Technology
ISBN: 0-08-0431526
pp. 5271–5274