Evaluation of Thermally Modified Wood by Means of Stress Wave and Ultrasound Nondestructive Methods

  • Cláudio Henrique Soares DEL MENEZZI Department of Forest Engineering Faculty of Technology University of Brasília
  • Marcela R. S. AMORIM Department of Forest Engineering Faculty of Technology University of Brasília
  • Mirian A. COSTA Department of Forest Engineering Faculty of Technology University of Brasília
  • Loureine R. O. GARCEZ Department of Forest Engineering Faculty of Technology University of Brasília
Keywords: acoustic methods, multiple regressions, mechanical properties

Abstract

The paper aimed at studying the potential of two nondestructive methods to estimate the wood mechanical properties and mass loss due to thermal treatments. In this study, a low-density tropical hardwood species Simarouba amara (marupá) was used. Forty small beams with dimensions of (25×25×400) mm (width×thickness×length) were cut from this species. Initially, the beams were nondestructively tested using stress wave and ultrasound methods. Stress wave velocity (Swv), ultrasound velocity (VLL), dynamic modulus of elasticity (Ed) and stiffness coefficient (CLL) were longitudinally determined. Afterwards, the beams were thermally treated using a chamber without air circulation under atmospheric pressure. Two schedules were tested: 160ºC for 180 minutes and 200ºC for 70 minutes. Mass loss (ML) due to thermal treatment was calculated and the thermally treated material was again nondestructively evaluated. Afterwards, modulus of rupture (fm), modulus of elasticity (EM) and parallel compression strength (fc,0) of treated material were assessed. Backward linear multiple regression analysis was run in order to estimate these properties. Parameters investigated through nondestructive testing (before and after treatment) and derivative variables were used as independent variables, totaling 12 variables. For both treatment schedules, all parameters related to nondestructive techniques were affected by the thermal treatment, thus acoustic velocities and stiffness values were improved. It was found that all evaluated properties of treated material could be modeled at a reasonable level (R2 = 0.392 to 0.874) depending on the nondestructive method and treatment schedule used. Nevertheless, ultrasound method fitted the most suitable models for a large number of properties. The utilization of variables from both methods together yielded better models whose R2 value ranged from 0.466 (fm) to 0.941 (EM). It was found that the most important nondestructive variables which entered into the models were: Swv before and after treatment, VLL after treatment, Ed before treatment and CLL after treatment. Finally, it could be concluded that stress wave and ultrasound nondestructive methods presented great potential to evaluate properties of thermally treated wood material.

DOI: http://dx.doi.org/10.5755/j01.ms.20.1.3341

Published
2014-03-23
Section
POLYMERS AND COMPOSITES, WOOD