Mode-I interlaminar fracture of carbon/epoxy cross-ply composites A.B. de Morais a, *, M.F. de Moura b , A.T. Marques b , P.T. de Castro b a University of Aveiro, Department of Mechanical Engineering, Campus Santiago, 3810 Aveiro, Portugal b University of Porto, Faculty of Engineering, Department of Mechanical Engineering and Industrial Management, Rua Dr. Roberto Frias, 4200-465 Porto, Portugal Received 13 December 2000; received in revised form 27 November 2001; accepted 30 November 2001 Abstract Mode-I double-cantilever beam (DCB) tests were performed on carbon/epoxy [0  /90  ] 12 specimens. The starter crack was created atmid-thickness,betweenthe0and90  mid-layers.Duringthetests,however,thecrackalsopropagatedalongtheneighbouring0  / 90  interface and within the 90  mid-layer. Nevertheless, the test results were apparently consistent with the assumptions of the corrected beam theory (CBT) that was used to obtain the interlaminar critical strain energy release rate, G Ic . The measured values were higher than those of unidirectional [0  ] 24 specimens, especially the final propagation values. A finite-element analysis con- firmed the applicability of the CBT for interlaminar propagation along the two 0  /90  interfaces. The results also indicated that the intralaminar G Ic is significantly smaller than the interlaminar G Ic . This will prevent pure interlaminar propagation in multi-direc- tional specimens with high interlaminar fracture toughness. # 2002 Elsevier Science Ltd. All rights reserved. Keywords: A. Polymer-matrix composites; B. Fracture toughness; C. Delamination; C. Transverse cracking; C. Finite-element analysis 1. Introduction It is well known that high-performance laminated composites are susceptible to delamination e.g. by low- velocity impact. This damage may significantly reduce the compressive strength by promoting delamination buckling phenomena. There has been considerable interest in the characterisation of the delamination resistance of laminates. The mode-I double cantilever beam (DCB) test (Fig. 1) is commonly used for that purpose. The test procedures are nowadays well estab- lished for unidirectional [0  ] n laminates e.g. in ISO/DIS 15024 [1]. The results, expressed in terms of the critical strain energy release rate, G Ic , are used mostly for com- parative purposes. In fact, practical applications usually involve multidirectional laminates, and it is observed that delaminations occur between layers of different orientations. It is, therefore, essential to study the interlaminar fracture of multidirectional specimens if critical strain energy release rates are to be applied in design. Various studies have already been conducted on the mode-I interlaminar fracture of multidirectional com- posites e.g. [2–10]. It is generally reported that G Ic values at initiation, and especially at propagation, are higher than those of [0  ] n laminates. The main feature seems to be that, after relatively small interlaminar propagation, the crack jumps to a neighbouring inter- face.Instudiesinvolvingspecimenswithinitialcracksin +45  /–45  interfaces, the intralaminar damage and the extensive fibre bridging behind the crack tip were responsible for a pronounced increase of G Ic [2,7–9]. Choi et al. [7] noted that this complex damage mor- phology resulted in significant stiffness variations throughout the test. Data reduction schemes based on the specimen flexural modulus should, therefore, be avoided. Moreover, the meaningfulness of the measured G Ic becomes obviously questionable. Schapery et al. [4] have suggested the use of the J integral to take into account the various damage mechanisms. In order to keep a pure interlaminar propagation, Robinson and Song [6] used a DCB specimen with pre- delaminatededges.Theyobservedthatcrackjumpingin specimens with +45  /+45  and +45  /45  interfaces could be suppressed if the pre-inserted edge delamina- tions were wide enough. In [7], however, it was not 0266-3538/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved. PII: S0266-3538(01)00223-8 Composites Science and Technology 62 (2002) 679–686 www.elsevier.com/locate/compscitech * Corresponding author. Tel.: 351-234-370830; fax: +351-234- 370953. E-mail address: abm@mec.ua.pt (A.B. de Morais).