PSI - Issue 10

E. Tsetsekou et al. / Procedia Structural Integrity 10 (2018) 227–234 E. Tsetsekou et al. / Structural Integrity Procedia 00 (2018) 000 – 000

229

3

according to the ASTM D905 (Fig.2). Three natural adhesives, casein, rabbit-skin glue and fish glue and two synthetic, a methyl acrylate and ethyl-methacrylate copolymer (Paraloid B72) and polyvinyl acetate glue (Ravemul M18- Vinavil) were applied based on the ASTM D905. Four replicates from each species of wood were used for each ad hesive. In total 120 pairs were assembled, 60 for each species. For the fish glue and rabbit-skin glue, aqueous solutions of 40% w/v have been prepared (Schellmann (2007)). Casein and Paraloid B72 were made in 50% w/v solutions in water and acetone respectively (Rivers and Umney (2003)), whilst Vinavil was a ready-made adhesive. The application of adhesives on both species of wooden samples was implemented on axial surfaces according to the ASTM D905. Pressure was then applied onto assembled samples for 72 hours. Post-cure conditioning was under taken i n a climate chamber (Binder KBF P 720) for 7 days under constant conditions (T=23± 2 o C, RH=55% ± 2).

Fig. 2 . Specimens’ dimensions and assembly position.

After curing, one group of 40 samples (20 for each species of wood) was aged with UV radiation under a XENON Arc lamp with a window glass filter, in an Atlas Suntest XLS+ chamber at an irradiance level of 350 W/m 2 . According to the ASTM G151, samples had to remain in the chamber until a distinct change in colour is produced. In order to determine the duration of the ageing cycle, five samples from both species of wood were coated with the five ad hesives and put under the XENON Arc lamp. Change in colour was observed after 98 hours when samples have been exposed to a total of 120960 kJ/m 2 . This duration of exposure was then followed for all the 40 assembled samples. A second group of 40 samples (20 for each species of wood) was subjected to ageing, with two different cycles of RH and temperature in a climate chamber Binder KBF P 720, based on the ASTM 1183. The duration of the first (T: 23± 2 o C and RH: 78% ± 2) and the second cycle (T: 48.5± 2 o C and RH : 23% ± 2) was 24 hours . Then the first cycle was repeated for 72 hours and upon completion, was followed by the 2 nd cycle conditions for 48 hours. Based on the standard this alternation of the first and second cycle was repeated for one more week. A last group of 40 samples was kept without being subjected to ageing, as control samples. The bond strength was assessed on aged samples and controls by shear stress by compression on an Instron 3300 machine with continuous motion of the movable head, at a rate of 5mm/min, to failure of bond, based on the ASTM D905 (Fig.3). All measurements taken were elaborated with the Blue Hill software. Colour stability under UV exposure in the Atlas chamber, was evaluated by a Lovibond RT Series SP60 colori meter, according to the European Standard EN 15886. The colour measurements were made on the coated samples used to determine the duration of the UV cycle. Both the colour of the bare wood surface and of the coated with adhesive surface, were measured before and after ageing. Reversibility of the adhesives was also evaluated before and after ageing with the use of solvents. Warm water was used for the removal of fish glue and rabbit-skin glue, water for casein and Vinavil and acetone for Paraloid B72. The results were visually examined.

3. Results and discussion

Results obtained on maximum load at the break point are presented in graphs (Figs.4-6). For maple controls, fish glue demonstrated the highest bond strength followed by Paraloid B72, rabbit-skin glue, Vinavil and lastly casein (Fig.4). For fir control samples, fish glue and Paraloid B72 again demonstrated the highest bond strength, however,