HYGROTHERMAL DEGRADATION OF TOUGHENED ADHESIVE JOINTS: THE .

HYGROTHERMAL DEGRADATION OF TOUGHENED ADHESIVE JOINTS:THE CHARACTERIZATION AND PREDICTION OF FRACTURE PROPERTIESbyAboutaleb AmeliA thesis submitted in conformity with the requirementsfor the degree of Doctor of PhilosophyGraduate Department of Mechanical and Industrial EngineeringUniversity of Toronto Copyright by Aboutaleb Ameli (2011)

HYGROTHERMAL DEGRADATION OF TOUGHENED ADHESIVE JOINTS: THECHARACTERIZATION AND PREDICTION OF FRACTURE PROPERTIESAboutaleb AmeliPh.D.Department of Mechanical and Industrial EngineeringUniversity of Toronto2011AbstractThe main objective of this work was to develop a framework to predict the fracturetoughness degradation of highly toughened adhesive joints using fracture test data obtained byaccelerated open-faced degradation method.First, the mixed-mode fracture resistance (R-curve) behavior of two rubber-toughenedepoxy-aluminum adhesive systems was measured and could be fit in a bilinear R-curve model.Then, open-faced DCB (ODCB) specimens of the same adhesive systems were aged over arelatively wide range of temperature, relative humidity (RH) and time, dried and tested tocharacterize the irreversible evolution of the mixed-mode fracture R-curves. The R-curvebilinear model parameters of adhesive system 1 varied significantly with degradation while thatof adhesive system 2 remained unchanged.The absorption and desorption of water in the adhesives cast wafers was measuredgravimetrically. The absorption data were fitted to a new sequential dual Fickian (SDF) modelwhile water desorption was modeled accurately using Fick’s law. A significant difference wasobserved between the amounts of retained water in the two adhesives after drying.ii

An exposure index (EI) was defined as the integral of water concentration over time andcalculated at all points in the ODCB and closed DCB joints. The fracture toughness of the closedjoints was then predicted from these calculated EIs by making reference to fracture toughnessdata from the ODCB specimens degraded to various EI levels. To verify the predictions, fractureexperiments and analyses were carried out for closed DCB joints. Good agreement was foundbetween the predicted and experimentally measured fracture toughness values for thedegraded closed DCB joints.Furthermore, the crack path and fracture surface characteristics were evaluated as afunction of the degree of aging using optical profilometery. The unexpected crack path in themixed-mode fracture of unaged open-faced DCB specimens was addressed. The results showeda strong relationship between fracture surface parameters and the critical strain energy releaserate, Gcs, irrespective of the type of adhesive and exposure condition.iii

Dedicated to my beloved, Nafiseh.iv

AcknowledgementsFirst and foremost, I would like to express my sincere gratitude and appreciation to mysupervisors, Professor Jan K. Spelt and Professor Marcello Papini for providing me with thecontinuous guidance, enthusiasm and encouragement to assist me in conducting a successfulresearch. Their visions, insights and suggestions not only fueled enormously my research butalso will have an everlasting influence in my professional career. I feel extremely honored andfortunate to have such supportive mentors.I would like to thank my Ph.D. committee members, Professor Anthony Sinclair andProfessor Hani Naguib for their valuable comments and suggestions offered during my annualexams.I am grateful of the financial support from General Motors of Canada, Ontario Centresof Excellence and Natural Sciences and Engineering Research Council of Canada. Regularcommunications and progress meetings with the members of General Motors Research andDevelopment and Planning was a valuable source of technical information and provided aninsight to the industry challenges. My special thanks at GM R&D and Planning goes to Dr.Jessica Schroeder, Dr. Douglas Faulkner, Dr. Justin Gammage (GM Canada), Dr. Blair Carlson andDr. John Ulicny.I would like to extend my acknowledge to the other members of the Materials andProcess Mechanics Laboratory, Amirhossein Mohajerani, Shahrokh Azari, Naresh Datla, SivaNadimpalli, Dwayne Shirely and Minseok Jhin for sharing and discussing all kinds of ideas andfor being always there whenever I needed help from them in anything.Last in this list but first in my heart is my wonderful wife, Nafiseh. I wish to express mywarmest gratitude to you for giving me endless support, love, caring and happiness. Withoutyou, I would have never been able to pursue my dreams. This work is dedicated to you Nafiseh,my angel of happiness.v

Table of ContentsAbstract . iiAcknowledgements . vTable of Contents . viList of Tables . xiList of Figures. xivChapter 1: Introduction .11.1. Motivation .11.2. Objectives .31.3. Organization of Thesis .41.4. References .6Chapter 2: Fracture R-curve characterization of toughened epoxy adhesives, .72.1. Introduction .72.2. Experimental procedures .82.2.1. Materials and joint fabrication .82.2.2. Fracture test methodology . 102.3. Fracture toughness calculations . 122.4. Results and discussion . 132.4.1. Validation of methods . 132.4.2. Characterization of R-curves . 162.4.3. The effect of adherend stiffness on the bilinear model parameters . 202.4.4. The effect of mode ratio on the R-curve bilinear model parameters . 212.4.5. The effect of bondline thickness on the R-curve bilinear model parameters . 222.4.6. The effect of initial geometry on the R-curve bilinear model parameters . 262.5. Conclusions . 292.6. Appendix. 302.7. References . 32Chapter 3: Hygrothermal properties of highly toughened epoxy adhesives, . 343.1. Introduction . 343.2. Mathematics of diffusion models. 363.2.1. Dual Fickian model . 363.2.2. Langmuir model. 39vi

3.2.3. Fickian model in desorption . 403.3. Experimental procedure . 403.4. Results and discussion . 423.4.1. Moisture absorption . 423.4.2. Moisture desorption. 543.4.3. XPS analysis . 603.5. Conclusions . 603.6. References . 62Chapter 4: Fracture R-curve of a toughened epoxy adhesive as a function of irreversible degradation, . 644.1. Introduction . 644.2. Experimental procedures . 664.2.1. Open-faced DCB specimen fabrication. 664.2.2. Gravimetric measurements and open-faced aging . 694.2.3. Fracture test methodology . 694.2.4. Measurement of residual adhesive thickness and surface roughness. 704.3. Results and discussion . 704.3.1. Water diffusion . 704.3.2. R-curve bilinear model . 724.3.3. Degradation of steady-state fracture toughness, Gcs . 734.3.4. R-curve degradation . 844.3.5. Effect of phase angle on degradation . 884.5. Conclusions . 894.6. References . 91Chapter 5: Hygrothermal degradation of two rubber-toughened epoxy adhesives: Application of openfaced fracture tests . 935.1. Introduction . 935.2. Experimental procedure . 945.2.1. Gravimetric measurements . 955.2.2. Dynamic mechanical thermal analysis . 965.2.3. Open-faced DCB specimen fabrication. 965.2.4. Fracture test methodology . 975.3. Results and discussion . 98vii

5.3.1. Water absorption and desorption . 985.3.2. XPS analysis . 1005.3.3. Dynamic mechanical thermal analysis . 1025.3.4. Fracture strength degradation . 1045.4. Conclusions . 1175.5. Appendix: Diffusion model . Error! Bookmark not defined.5.6. References . 119Chapter 6: Crack path selection in the fracture of fresh and degraded epoxy adhesive joints . 1226.1. Introduction . 1226.2. Experimental procedures . 1246.2.1. Materials and DCB specimen preparation. 1256.2.2. Open-faced specimen preparation. 1256.2.3. Fracture test procedure. 1266.2.4. Measurement of fracture surface profiles and crack path . 1266.3. Finite element modeling . 1266.4. Plastic zone evolution with crack growth (R-curve behavior) . 1296.5. Steady-state plastic zone and phase angle . 1346.6. Crack path . 1366.6.1. Crack path in fresh joints under mode-I . 1366.6.2. Crack path in fresh joints under mixed-mode loading . 1396.6.3. Crack path in degraded joints. 1456.7. Conclusions . 1506.8. References . 152Chapter 7: Evolution of crack path and fracture surface with degradation in rubber-toughened epoxyadhesive joints: Application to open-faced specimens . 1557.1. Introduction . 1557.2. Experimental procedure . 1577.2.1. Materials. 1587.2.2. ODCB specimen fabrication . 1587.2.3. Fracture test procedure. 1597.2.4. Measurement of fracture surface profiles and crack path . 1607.2.5. Residual stress measurement . 161viii

7.3. Results and discussion . 1617.3.1. Fresh ODCB joints – unexpected crack path . 1617.3.2. Aged ODCB joints. 1717.3.3. Relation between crack path and plastic zone . 1797.4. Conclusions . 1817.5. References . 183Chapter 8: Prediction of environmental degradation of closed adhesive joints using data from openfaced specimens . 1868.1. Introduction . 1868.2. General framework . 1878.3. Experimental procedure . 1898.3.1. Gravimetric measurements . 1908.3.2. ODCB preparation . 1908.3.3. CDCB preparation . 1908.3.4. Fracture testing .

iii An exposure index (EI) was defined as the integral of water concentration over time and calculated at all points in the ODCB and closed DCB joints. The fracture toughness of the closed joints was then predicted from these calculated EIs by making reference to fracture toughness data from the ODCB specimens degraded to various EI levels. To verify the predictions, fracture