Thickness influence on a structural methacrylate adhesive behavior
25
0
0
Texte intégral
(2)
(3)
(4)
(5)
(6)
(7)
(8) . an author's
(9)
(10)
(11) https://oatao.univ-toulouse.fr/19884
(12)
(13) .
(14) . Jaillon, Agathe and Jumel, Julien and Lachaud, Frédéric and Paroissien, Eric and Renart, Jordi Thickness influence on a structural methacrylate adhesive behavior. (2018) In: 6th World Congress on Adhesion and Related Phenomena (WCARP 2018), 25 February 2018 - 1 March 2018 (San Diego, United States). (Unpublished). .
(15)
(16)
(17)
(18)
(19)
(20)
(21)
(22)
(23) .
(24) WCARP 2018 – February 28. Thickness influence on a structural methacrylate adhesive behavior Presented by: Agathe Jaillon1 • agathe.jaillon@isae.fr • +335 61 17 12 14. Co-authors: Julien Jumel2, Frédéric Lachaud1, Eric Paroissien1, Jordi Renart3 1. 2. 3.. Institut Clément Ader (ICA), Université de Toulouse, CNRS, INSA, ISAE-SUPAERO, MINES ALBI, UPS, F-31400 Toulouse, France Université de. Bordeaux, Arts et Metiers ParisTech, CNRS, I2M, UMR 5295, F-33400 Talence, France Universitat de Girona · Department of Mechanic Engineering and Industrial Construction · AMADE, Spain · Girona. www.institut-clement-ader.org. Toulouse, Albi, Tarbes.
(25) Context • Global project To develop an adhesive bond manufacturing process which ensure its reliability through supervision and numerical simulation of the part life cycle. Methacrylate structural adhesive. Optimal design. Supervision. Charactérization Characterization. Conformity control. Automated manufacturing. • Scientific problematic •. Characterization approaches for the evaluation and modeling of adhesive thickness influence on adhesive bond failure behavior www.institut-clement-ader.org. Toulouse, Albi, Tarbes. 2.
(26) Content • Context • Experimental studies • • •. Bulk adhesive tests Arcan tests DCB tests. • Numerical modeling of a DCB test • Comparison between experimental and numerical test results • Conclusions and perspectives. www.institut-clement-ader.org. Toulouse, Albi, Tarbes. 3.
(27) Experimental Study. Numerical Analysis. Results Comparison. Conclusions. Bulk adhesive manufacturing • Adhesive specification • • •. Methacrylate structural adhesive SAF30MIB (Bostik) Bi composants thermoplastic adhesive in paste Polymerization at room temperature for 24h. Cured specimen. • Manufacturing protocol • • •. PTFE mould Specimen’s size 150x10x4mm [NF ISO 527-2] Bulk specimen manufacturing Recovered by a polypropylene film Inhibited by oxygen. www.institut-clement-ader.org. Toulouse, Albi, Tarbes. 4.
(28) Experimental Study. Results Comparison. Numerical Analysis. Conclusions. Bulk adhesive tensile tests • Test condition • • •. Monotonic test on an Instron 100kN Speed rate : 5mm/min Measurements made by 3D digital image correlation (DIC). 14. Stress (MPa). 12 10 8. E (MPa) 614 (±143). 6 4. υ(-) 0.37 (±0,02). Smax (MPa) 11.8 (±1,6). 2 0 0%. 5%. 10%. 15%. 20%. 25%. 30%. 35%. Strain ( % ). www.institut-clement-ader.org. Toulouse, Albi, Tarbes. 5.
(29) Experimental Study. Numerical Analysis. Results Comparison. Conclusions. ARCAN specimen manufacturing • Specimen specifications •. Massive adherends • Round beaks [Cognard2005]. •. Bonded surface : • 21x40mm. •. Surface preparation : • degreasing with isopropanol. •. Calibrated thicknesses : • 0.2mm and 1mm. •. Excess adhesive removed with a cylindrical spatula. www.institut-clement-ader.org. Toulouse, Albi, Tarbes. 6.
(30) Experimental Study. Numerical Analysis. Results Comparison. Conclusions. ARCAN specimen manufacturing • Specimen specifications •. Massive adherends • Round beaks [Cognard2005]. •. Bonded surface : • 21x40mm. •. Surface preparation : • degreasing with isopropanol. •. Calibrated thicknesses : • 0.2mm and 1mm. •. Excess adhesive removed with a cylindrical spatula. www.institut-clement-ader.org. Toulouse, Albi, Tarbes. 7.
(31) Experimental Study. Numerical Analysis. Results Comparison. Conclusions. ARCAN specimen manufacturing • Specimen specifications •. Massive adherends • Round beaks [Cognard2005]. •. Bonded surface : • 21x40mm. •. Surface preparation : • degreasing with isopropanol. •. Calibrated thicknesses : • 0.2mm and 1mm. •. Excess adhesive removed with a cylindrical spatula Thickness calibrator. www.institut-clement-ader.org. Positioning axis. Toulouse, Albi, Tarbes. 8.
(32) Experimental Study. Numerical Analysis. Results Comparison. Conclusions. ARCAN specimen manufacturing • Specimen specifications •. Massive adherends. Adhesive. • Round beaks [Cognard2005]. •. Bonded surface : • 21x40mm. •. Surface preparation : • degreasing with isopropanol. •. Calibrated thicknesses : • 0.2mm and 1mm. •. Excess adhesive removed with a cylindrical spatula. www.institut-clement-ader.org. Toulouse, Albi, Tarbes. 9.
(33) Experimental Study. Results Comparison. Numerical Analysis. Conclusions. validation of specimen geometry – FE model • Stress distribution. • Stress at the edge. 700. 1000. Curved adhesive. 800. Straight adhesive. Adherend Adhesive Adherend Adhesive. 600 400 200. Von Mises Stress (MPa). Von Mises stresses (MPa). 1200. 0. 600 500 400 300. No Beak Triangular beak Round Beak. 200 100 0. no beak. •. L=15.7mm. triangular beak. 0. round beak. Advantages of curved adhesive • Decrease stress concentration. 5. 10. 15. 20. Abscissa along the overlap (mm). •. Advantages of round beaks • Decrease edge’s stress concentration • Increase length where stress is constant. www.institut-clement-ader.org. Toulouse, Albi, Tarbes. 10.
(34) Experimental Study. Numerical Analysis. Results Comparison. Conclusions. ARCAN tests Tension. • Test conditions • • • • •. Modified Arcan on Zwick 10kN Digital image correlation on the adherends Speed rate for t = 0.2mm: 0.1mm/min Speed rate for t = 1mm: 0.2mm/min Loadings : tension, shear and tension-shear. • Matlab analysis DIC analysis. α. Shear. Arcan experimental set-up. Image correction. Results. Fixed substrat. Movable substrat. www.institut-clement-ader.org. Toulouse, Albi, Tarbes. 11.
(35) Experimental Study. Numerical Analysis. Results Comparison. Conclusions. ARCAN Results • Tension: α=0° t=1mm. 14. t=0.2mm. Stress (Mpa). 12 10 8 6 4. Tension. 2 0 0. 0.1. 0.2. Strain ( % ). 0.3. 0.4. t (mm). Yt (MPa). Smax (MPa). 0.253(±0.06) 0.796 (±0.04). 1650(±850) 1750 (unk.). 10.56 (±0,24) 13.07 (±0,14). www.institut-clement-ader.org. Toulouse, Albi, Tarbes. 12.
(36) Experimental Study. Numerical Analysis. Results Comparison. Conclusions. ARCAN Results • Tension: α=0° t=1mm. 14. t=0.2mm. Stress (Mpa). 12 10 8 6 4. Tension-shear. 2 0 0%. 10%. 20%. 30%. Strain ( % ). • Shear α=90°. t=1mm. 14. 40%. t=0.2mm. Stress (MPa). 12 10. t (mm). 45° Smax (MPa). εr (%). 0.183(±0.01) 0.765 (±0.21). 12,51 (±0,11) 13.21 (±0,10). 250 (±54) 119(±8). 8 6 4 2 0 -5%. 45%. 95%. 145%. Strain (%). t=1mm. 195%. 245%. 295%. www.institut-clement-ader.org. Toulouse, Albi, Tarbes. 13.
(37) Experimental Study. Results Comparison. Numerical Analysis. Conclusions. ARCAN Results • Tension: α=0° t=1mm. 14. t=0.2mm. Stress (Mpa). 12 10 8 6 4. Tension-shear. 2 0 0%. 10%. 20%. 30%. Strain ( % ). • Shear α=90°. t=1mm. • Failure envelope. t=0.2mm. 14. Tensile stress (MPa). 14. 40%. Stress (MPa). 12 10 8 6 4 2. 0°. 12. t=0.2mm t=1mm. 45°. 10 8 6 4. 90°. 2 0. 0 -5%. 45%. 95%. 145%. Strain (%). t=1mm. 195%. 245%. 295%. 0. www.institut-clement-ader.org. 5. 10. Shear stress (MPa). Toulouse, Albi, Tarbes. 15. 14.
(38) Experimental Study. Numerical Analysis. Results Comparison. Conclusions. DCB specimen manufacturing • Specimen specification • • • •. Adherends’ size : 195x15x10mm Overlap length : 145mm Initial crack : 35mm PTFE film Calibrated thickness : • 0.2mm and 1mm. DCB adherends with PTFE film for thickness calibration. DCB specimen assembly tool. www.institut-clement-ader.org. Toulouse, Albi, Tarbes. 15.
(39) Experimental Study. Numerical Analysis. Results Comparison. Conclusions. DCB tests • Test conditions • • • •. On a Zwick 10kN Speed rate: 2mm/min Digital image correlation on both sides Cohesive failure DCB experimental set-up. Adherend displacement. Crack tip. Adherend rotation. Openning measurement. initial crack tip taken up close. Full DCB specimen. www.institut-clement-ader.org. Toulouse, Albi, Tarbes. 16.
(40) Experimental Study. Results Comparison. Numerical Analysis. Conclusions. DCB results • Load displacement curves •. 1400. Hypothesis :. propagation initiation. 1200. • No plastic strain in adherends • Adhesive behavior not time dependent. Load (N). 1000. F. t=1mm. 800 600. t=0.35mm. 400. θ. Adhesive. Adherends. 200 0. h t. 𝛿. 0. 5. 10. 15. Opening at loading point δ (mm). θ F. 𝑎0. •. 𝐿 Schematic representation of a DCB test. Comments: • Modulus not thickness-dependent • Crack propagation thickness-dependent. www.institut-clement-ader.org. Toulouse, Albi, Tarbes. 17.
(41) Experimental Study. Results Comparison. Numerical Analysis. Conclusions. DCB results • Energy release rate •. • Traction-separation law •. J integral computation. 𝟐𝑭𝜽 [Andersson2004] 𝑱= 𝒃. 5. 𝒅𝑱 𝝈= 𝒅𝜹. 12. 4.5. t=1mm. 3.5 3. GIc. 2.5. 2 1.5. t=0.35mm. 1. Model DCB 7 DCB 5 DCB 1. 10. Stress (MPa). 4. J (N/mm). Stress computation for t=0.2mm. 8 6 4 2. 0.5 0. 0 0. 5. 10. 15. 0%. 20%. GIc (J/m²) 1730 (±347) 4270 (±180). 60%. 80%. GIc (J/m²) 1691 (±99). 𝜺𝒑 (%) 82 (±19). Strain ( % ). Opening at loading point δ (mm). t (mm) 0.35 (±0,06) 1.31 (unk.). 40%. t (mm) 0.35 (±0,06). Yt (MPa) 762 (±162). www.institut-clement-ader.org. Smax (MPa) 9.4 (±1,3). Toulouse, Albi, Tarbes. 18.
(42) Experimental Study. Results Comparison. Numerical Analysis. Conclusions. Strain-stress curves comparison 14. • Comments. 12. • • •. Stress (MPa). 10 8 6 4. •. DCB Model Arcan test Uni-axial tensil test. 2 0 0%. 10%. 20%. 30%. 40%. 50%. 60%. 70%. •. Strain (%). Test Tensile DCB ARCAN. t (mm) 3,96 (±1.6) 0.35 (±0,06) 0.253 (±0.06). Yt (MPa) 614 (±143) 762 (±162) 1650 (±850). Smax (MPa) 11.8 (±1,6) 9.4 (±1,3) 10.56 (±0,24). A (MPa) 3.93 (±0.12) 4.83 (±0.8) 2.57 (±1.8). •. 3 kind of tests with different stress state All tests have similar tendencies Modulus of the linear part are similar for DCB an tensile test Arcan tests have a significantly higher Modulus DCB and ARCAN tests have a similar plastic plateau Softening part only measurable during DCB test Which experimental results should be used for the numerical modeling of the adhesive failure behavior ?. www.institut-clement-ader.org. Toulouse, Albi, Tarbes. 19.
(43) Experimental Study. Results Comparison. Numerical Analysis. Conclusions. DCB models • Macro element [Lelias2015] • •. 1D Implemented in scilab 200 BBE in overlap length. 1st BBE Beam Element. • Cohesive zone model • •. ith BBE. Bonded Beam Element (BBE). Bilinear law as first approximation Trapezoid law identified in DCB tests. Stress. Stress. Damage initiation. Smax. (σ, ε). Yt. (σ, ε). Smax Damage propagation. Damage propagation. Yt. GIc. 𝜀𝑖. Damage initiation. 𝜀𝑝. Strain. GIc. 𝜀𝑖. www.institut-clement-ader.org. 𝜀𝑠 𝜀𝑝 Toulouse, Albi, Tarbes. Strain. 20.
(44) Experimental Study. Results Comparison. Numerical Analysis. Conclusions. Numeric VS experimental • Traction-separation law. • Load-openning curves. 900 1200 800. 9. t = 0.2 mm. 8. Bilinear. Bilinear. 1000 700. 7. Trilinear. Trilinear. 600 800. 6. t = 1 mm. 5 4. Load (N). Stress (MPa). 10. 500 600 400. t = 0.2 mm EXP Bilinear CZM Trilinear CZM. 300 400. 3. 1. 200 200 100. 0. 0. 2. 0. 25. 50. 75. 100. 125. 0. Yt (MPa) 710 710. Smax (MPa) 9.5 9.5. GIc (J/m²) 1950 4000. 2. 3. 4. Openning at loading point (mm). Strain (%). Adhesive thickness (mm) 0.35 1.31. 1. t = 1 mm EXP Bilinear CZM Trilinear CZM. • • •. Propagation correctly predicted Stiffness correctly predicted Inverse method to find CZM for 1mm-bond. www.institut-clement-ader.org. Toulouse, Albi, Tarbes. 21.
(45) Experimental Study. Numerical Analysis. Results Comparison. Conclusions. Conclusions & prospects • Experimental results •. Characterization of the adhesive failure behavior • 3 differents tests : tensile on bulk adhesive, ARCAN and DCB • 2 thicknesses: 0.2mm and 1mm • Identifiaction of different stress-strain curves dependant on the stress state of the test. •. Adhesive thickness influence on the failure behavior ? • Increase of the energy release rate GIc • Increase of plastic plateau value • Increase of strain at failure in shear. • Numerical modeling • •. Implementation of a bilinear and trilinear TS law in a ME model TS law identified for 0.2mm-bond not directly usable for 1mm-bond. • Perspectives • • •. Experimental Identification of 1mm to 5mm-bond traction separation law Sensivity study of CZM parameters Implementation of a elasto-viscoplatic cohesive element. www.institut-clement-ader.org. Toulouse, Albi, Tarbes. 22.
(46) Thank you for your attention. www.institut-clement-ader.org. Toulouse, Albi, Tarbes.
(47) References 1. 2. 3.. 4.. Norme, N.F EN ISO 527-2, Plastiques-Détermination des propriétés en traction-Partie 2: Conditions d'essai des plastiques pour moulage et extrusion, 2012 Cognard, J. Y.; Davies, P.; Gineste, B. & Sohier, L. Development of an improved adhesive test method for composite assembly design, Composites Science and Technology, 2005, 65, 359-368 Andersson, T. & Stigh, U. The stress--elongation relation for an adhesive layer loaded in peel using equilibrium of energetic forces, International Journal of Solids and Structures, 2004, 41, 413-434 Lelias, G.; Paroissien, E.; Lachaud, F.; Morlier, J.; Schwartz, S. & Gavoille, C. An extended semianalytical formulation for fast and reliable mode I/II stress analysis of adhesively bonded joints, International Journal of Solids and Structures, Elsevier, 2015, 62, 18-38. www.institut-clement-ader.org. Toulouse, Albi, Tarbes. 24.
(48)
Documents relatifs
In order to better understand the instantaneous mechanical behaviour of the claystone at different sample scales, we conducted three series of measurements for this study:
We propose a variational formulation of a macroscopic model for crowd mo- tion involving a conservation law describing mass conservation coupled with an eikonal equation giving the
L’accès à ce site Web et l’utilisation de son contenu sont assujettis aux conditions présentées dans le site LISEZ CES CONDITIONS ATTENTIVEMENT AVANT D’UTILISER CE SITE WEB.
(a) Transient response obtained for photochemical vapor generation of As from solutions spiked with 5 ng/mL As(III) and containing low molecular weight organic acids: I, 4.6 M
L‟ensemble des résultats d‟analyse tendent à montrer que les valeurs obtenues pour les critères décrivant la qualité physicochimique des différents laits
The rat uterine mast cells preincubated with oestradiol and then challenged with human EHRF gave histamine release values which were two- to threefold higher than values
European isolates of the Microsporidia Nosema apis and Nosema ceranae have similar virulence in laboratory tests on European worker honey bees... European isolates of the
Therefore, method variance is an instance of construct irrelevant variance (Messick, 1989) and a source of test invalidity. 156) stated that “When test performance is unduly
