Investigation of the Impact of Increasing Design Loads of a Conventional Structural Silicone Joint Using Finite Element Analysis and Hyperelastic Material Properties

• Published in: Durability of Building and Construction Sealants and Adhesives, 5th Volume - (STP 1583) pp. 217 - 234
• Main Authors: Kimberlain, Jon, Knowles, John A.
• Format: Book Chapter
• Language: English
• Published: 100 Barr Harbor Drive, PO Box C700, West Conshohocken, PA 19428-2959 ASTM International 22.06.2015
• Subjects: Adhesives & Sealants; Adhesives, Coatings, Sealants & Inks; Civil Engineering; Civil Engineering & Construction Materials; Curtain Wall; Finite Element Modeling; Hyperelastic; Mooney-Rivlin; Sealant Design Strength; Structural Sealant Glazing; Structural Silicone
• ISBN: 0803176112; 9780803176119
• DOI: 10.1520/STP158320140063

Design methodology for structural silicone glazed curtain wall assembly has been well established for many decades with proven performance of building projects around the globe. At its core, the design methodology is based on the individual performance of the materials, and the assembly is designed using physical and geometric principles to perform within an expected load and deflection which usually are confirmed via experiential confirmation via mock-up assemblies. Structural silicone performance and allowable design stress is based on physical relationships of stress due to the strain of a proto-typical joint. There has been recent interest in exploring increased or modified design stress of sealants based on similar methodology. However, the challenge in the methodology is that silicone materials behave as hyperelastic elastomers which indicate that the relationship of stress versus strain is not linear and the materials behave with different moduli depending on the location of the allowable stress on the stress–strain curve. Using finite element models, the paper presents a comparative analysis of a structural silicone joint designed via the conventional hand calculation at different design loads. The analysis illustrates the higher potential stress peaks developed within the sealant during loading versus the expected stress from the conventional method of calculation. Frame deflection and glass thickness were also factored to understand the variables’ impact to the distribution of stresses in the sealant joint compared to the original stress predicted from conventional methods of design. Methods to calibrate the actual material model used in the finite element software for hyperelastic materials are also presented. Physical assemblies were tested to compare the results of the software model, which also includes a modified test procedure to better understand the durability of the materials due to cyclical testing.