PSI - Issue 37

Almudena Majano-Majano et al. / Procedia Structural Integrity 37 (2022) 492–499 Almudena Majano-Majano et al. / Structural Integrity Procedia 00 (2019) 000 – 000

493

2

Keywords: Eucalyptus globulus; PUR; bonded joints; cohesive law; strain energy release rate; DCB; DIC

1. Introduction Hardwood species are gaining popularity in the European engineered timber market. Consequently, there is constant research in the development of new wood-based structural products, many of which use adhesive bonding for their production, such as glulam, cross laminated timber or laminated veneer lumber (e.g. Aicher et al. (2014), Ehrhart et al. (2016), Luedtke et al. (2015), Van Acker (2021)). In the development of glued laminated products using high density hardwood species, bonded joints (e.g. finger joints) could be a weak point because most adhesives are designed for softwoods. Specific standards for hardwood laminated products are currently under development. A better understanding of the behavior and properties of both the adhesive and the materials involved is therefore essential to produce reliable hardwood structural products and safe applications. Eucalyptus globulus L. is a temperate climate hardwood commonly used in the paper industry, but it stands out as a species of great potential for engineered timber products. It is one of the three hardwoods in Europe with the highest mechanical properties, being classified as D40 like beech and ash (EN 1912:2012). This feature, together with its great natural durability, makes this species the subject of recent research (e.g. Derikvand et al. (2019), Franke and Marto (2014), Majano-Majano et al. (2019, 2020)). In the field of adhesives, one-component polyurethane (1C-PUR) has become very usual for engineered timber products due to the existence of its own European classification standard and requirements (EN 15425:2008). It also offers advantageous properties, such as being free of formaldehyde and providing fast curing at room temperature. Recent studies show its potential when used with Eucalyptus globulus L (e.g. López-Suevos and Richter (2009), Lara Bocanegra et al. (2017, 2020)). Numerical modelling by finite element analysis is recognized as a useful tool for addressing structural safety and efficiency in the design of bonded joints. In this respect, the implementation of cohesive zone models (CZM) based on fracture mechanics seem to be suitable to deal with damage growth. In the definition of these models, one of the key points is to know the cohesive law that describes accurately the fracture process within the bond lines. The cohesive law relates the stresses and relative displacements between the glued faces ahead of the crack tip and can be determined by direct experimental methods. In wood bonded joints involving a mode I loading (tension), one of the most common methodologies applied for fracture characterization is the Double Cantilever Beam (DCB) test (e.g. Gagliano and Frazier (2001), Lavisci et al. (2003), Silva et al. (2013), Veigel et al. (2012), Xavier et al. (2011)). In the process, it is required to quantify the strain energy release rate ( G ) that initiates and propagates the crack during the test, which is an important property of the adhesive bond and can be derived by different compliance-based methods. The aim of the present work is to determine the cohesive law, using exclusively experimental data, of 1C-PUR bonded eucalyptus specimens under mode I loading from DCB tests. The evolution of strain energy release rate in the course of the tests is determined from the Resistance curves ( R -curves) by applying the Compliance Based Beam Method (CBBM) (de Moura et al. (2008a)), which is based on an equivalent crack length concept that avoids crack growth measurements during testing. The digital image correlation (DIC) technique is used to measure full-field opening displacements produced at the crack tip. Finally, by integrating these independent measurements, a direct evaluation of the cohesive law is proposed. To the best of the author’s knowledge, there are no studies on mode I fracture properties of eucalyptus bonded joints or other species bonded with 1C-PUR applying this data reduction methodology.

2. Materials and Methods 2.1. Specimen preparation

Eucalyptus globulus Labill wood from Galicia (Northwest Spain) was used in the present work. The specimens were cut from approximately knot-free boards (knot diameter less than 1/20 times the board width), which is a characteristic feature of this species. These boards showed an average longitudinal modulus of elasticity E L =19640