Shaped ductile ®bers to improve the toughness of
epoxy-matrix composites
Robert C. Wetherhold *, Foo Kwong Lee
Department of Mechanical and Aerospace Engineering, State University of New York at Bualo, Bualo, NY 14260-4400, USA
Received 16 May 2000; received in revised form 24 August 2000; accepted 10 October 2000
Abstract
Pull-out tests on ductile ®bers embedded in epoxy resin have shown an additional component of pull-out work from the plastic
deformation of the ®ber, which component is not available in traditional brittle ®bers. This additional deformation energy is
available to improve the toughness of composite materials without sacri®cing composite stiness. The plastic deformation energy in
ductile copper and nickel ®bers is exploited by anchoring the ®bers in the matrix through modi®cation of ®ber ends. Shaped ®ber
ends were produced by end-impacting and knotting ®bers to facilitate anchoring, similar to previous work with `bone-shaped' short
(BSS) ®bers. Both shaped and straight ®bers were embedded in epoxy to various depths and angles, and pull-out tests were per-
formed to determine whether an increase in pull-out work was evident for shaped ®bers. Cold-working the ®bers or treatment with
release agent prior to embedment served as a tool to isolate the frictional component from the plastic deformation component of
pull-out work during pull-out tests. The eect of ®ber yield stress and hardenability on the pull-out work was also investigated.
Calculations based on the test results and a model for predicting fracture toughness increment showed a 5±109% higher toughness
increment for shaped ®bers compared to straight ®bers, dependent on material conditions and ®ber orientation. # 2001 Published
by Elsevier Science Ltd. All rights reserved.
Keywords: Fracture toughness; Bridging; Shaped ductile ®bers; Single ®ber pull-out test
1. Introduction
Composites reinforced with discontinuous, randomly
oriented ®bres are comparatively easier and cheaper to
fabricate than their continuous ®ber counterparts. If the
need arises, ¯ow-induced ®ber orientation can help
achieve desired properties, oering ¯exibility in design.
Randomly distributed ®bers also result in better
mechanical properties in transverse directions, improving
the composite's performance in complex stress situations.
However, because of the short ®ber lengths and dis-
continuities at the ®ber ends, discontinuous-®ber-
reinforced composite materials are somewhat inferior to
their continuous-®ber counterparts when it comes to
strength and stiness. Also, fracture toughness is not as
high on account of the discontinuity of the reinforce-
ment.
Since the 1970s, many researchers have been trying to
improve upon the properties oered by discontinuous-®ber
reinforcements. The bulk of the research was on
improving the characteristics of the ®ber/matrix inter-
face, e.g. chemical coupling agents for glass ®bers and
oxidative treatments for carbon ®bers [1]. High ®ber/
matrix interface friction improves stiness enhancement
but does little to improve, or may even degrade, fracture
toughness. Low ®ber/matrix interface friction allows for
debonding and thus reduces the stress intensity at the
tip of an advancing crack in the matrix, improving
fracture toughness of the composite. However, the low
friction decreases strength and stiness enhancement.
This competing nature of the ®ber/matrix interface
properties has provided incentive for further work.
In the mid to late 1970s, Morton and Groves [2,3],
and Bowling and Groves [4,5] utilized ductile metal
®bers to reinforce a brittle matrix. Bowling and Groves
developed a model [4] for depicting the behavior of a
ductile ®ber bridging a crack in a brittle matrix. This
model provides a partial solution to the competing nature
described above because of the ability of the ductile
®ber to plastically deform as it bridges the crack. This
plastic deformation in the ®ber longitudinal direction
0266-3538/01/$ - see front matter # 2001 Published by Elsevier Science Ltd. All rights reserved.
PII: S0266-3538(00)00217-7
Composites Science and Technology 61 (2001) 517±530
www.elsevier.com/locate/compscitech
* Corresponding author. Fax: +1-716-645-3875.
E-mail address: mecrcw@acsu.bualo.edu (R.C. Wetherhold).