JOURNAL OF MATERIALS SCIENCE 26 (1991) 6565-6575 Tensile failure mechanisms in synthetic fibre-reinforced mortar YOUJIANG WANG School of Textile and Fiber Engineering, Georgia Institute of Technology, Atlanta, Georgia 30332, USA V.C. LI Department of Civil Engineering, University of Michigan, Ann Arbor, Michigan 48109, USA S. BACKER Department of Mechanical Engineering, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139, USA The ultimate tensile behaviour of fibre-reinforced cementitious composites is closely related to its failure mechanisms which in turn are dependent on reinforcement parameters such as fibre characteristics and the fibre/matrix interface properties. Based on the direct tensile tests of mortar specimens reinforced with various synthetic fibres, this paper attempts to explain such relationships and to indicate directions towards more effective fibre reinforcement. 1. Introduction Concrete, a ubiquitous construction material, has low tensile strength and low fracture toughness which limit its use in many critical applications. Research has indicated that the tensile behaviour of concrete can be effectively upgraded by fibre reinforcement at rela- tively low fibre volume fractions, typically 0.5% to 3%. Because of such improvements, fibre-reinforced cementitious composites (FRC) are suitable for many applications. The tensile behaviour of FRC is a fundamental material property which can be characterized by means of a stress-crack opening width (~-8) curve [1]. In an experimental study of fibre reinforcement of mortar with various synthetic fibres [2], it was ob- served that the measured or-6 curve was strongly influenced by the fibre reinforcement, particularly by the fibre types used, and that, as the differences in ~-6 curves imply, some reinforcements were not as effec- tive as others. Here, the tensile failure mechanisms in mortar spe- cimens reinforced with various synthetic fibres are studied, based on observations under an optical microscope as well as under a scanning electron microscope (SEM). These mechanisms are related to both the tensile behaviour (or-6 curves) of the com- posites and the reinforcement parameters (fibre type, size and volume fraction). The purpose of this study was to point out directions towards optimization of FRC properties with respect to the reinforcement parameters. 2. Direct tensile testing of FRC Mortar matrix, composed of Type III cement, mortar sand and water (weight ratio = 1:1:0.5) was reinfor- ced with various synthetic fibres, including Kevlar 49 (aramid, DuPont Company), Technora (aramid, Teijin Ltd), Spectra 900 (high strength polyethylene, Allied Corporation), and Herculon PP (undrawn polypro- pylene, Hercules Inc.). Notched specimens, such as shown in Fig. 1, were tested in direct tension using the test fixture illustrated in Fig. 2. A stress-crack opening width (~-6) curve was recorded for each test, with the specimen load measured by the machine load cell, and the specimen crack width monitored by a pair of linear variable differential transformers (LVDTs) attached to the specimen. Details on materials, mixing, curing, and testing of FRC specimens can be found elsewhere [-2, 3, 4]. Average stress-crack opening width (cy-~) curves are shown in Fig. 3 for the synthetic FRC tested, along with the information on fibre length (Lf), diameter (dr) and volume fraction (V0. Tensile strengths and fracture energies given by the areas under the or-6 curves are summarized in Table I. g 76.2 50.8 76.2 Notched cross-section Figure 1 Dimensions (mm)of direct tensiletest specimens. 0022-2461/91 $03.00 + .12 9 1991 Chapman & Hall 6565