Mechanical performance of glulam products made with Portuguese poplar

• Published in: European journal of wood and wood products Vol. 78; no. 5; pp. 1007 - 1015
• Main Authors: Monteiro, Sandra R. S., Martins, Carlos, Dias, Alfredo M. P. G., Cruz, Helena
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.09.2020; Springer Nature B.V
• Subjects: Bend strength; Biomedical and Life Sciences; Ceramics; Composites; Forest products; Glass; Glulam; Hardwoods; Laboratories; Life Sciences; Machines; Manufacturing; Mathematical models; Mechanical properties; Natural Materials; Numerical models; Numerical prediction; Original; Poplar; Processes; Species; State-of-the-art reviews; Three dimensional models; Timber; Timber (structural); Wood products; Wood Science & Technology
• ISSN: 0018-3768; 1436-736X
• DOI: 10.1007/s00107-020-01569-y

Due to its average mechanical properties, poplar, a fast-growing species, has been disfavored compared to stronger species for several decades. Wood-based products may help changing that perspective and thus, poplar has been gaining its market share for structural uses. A state-of-the-art review concerning the use of poplar to produce glued laminated products, with special focus on the use of Portuguese poplar, is presented. The Portuguese forest produces a great variety of species. The most common poplar species found in this country are Populus x canadensis , P. nigra L., and P. alba L. Despite its limited availability, and the market hesitation on its structural application, recent studies on poplar grown in the Portuguese forest showed its suitability for structural purposes. Glued laminated timber (GLT) beams made with this species revealed a very promising mechanical behavior. Bending strength tests evidenced a ductile behavior on more than 70% of the beams, which motivated deepening the study on the raw material used to produce those beams. To predict the mechanical behavior of such beams, a 3D numerical model was developed. The numerically predicted results were compared with the experimental ones, showing very good agreement between both approaches.