Date of Award


Degree Type


Degree Name

Master of Science


Civil Engineering

Major Professor

Z. John Ma

Committee Members

Richard Bennett, Dayakar Penumadu


At the present time a large portion of America’s bridge infrastructure is aging, and in many cases approaching its design live. Therefore, there is a need for structurally sound, low-maintenance, and rapidly constructible alternatives for conventional materials. Due to this need there has been a renewed interest in structural glass fiber reinforced polymer (GFRP) products. A large portion of the US is regularly subjected to freezing temperatures and therefore the numerous gaps in our knowledge and understanding of the low-temperature response of GFRP materials need to be addressed. Although strides have been made in researching the cold climate responses of GFRP materials, a lack of comparable testing procedures and reproducible results has lead to confusion and a lack of confidence applying them. Therefore this thesis presents a detailed theoretical analysis of a structural GFRP bridge deck panel, and documents a research program that observed the effects of low-temperature and strain levels on the longitudinal modulus of GFRP samples. A series of GFRP coupon laminates were tested at 1000, 2000, and 3000 micro-strain levels at temperatures down to –31 F [-35 C]. Both biaxial and uniaxial samples subjected to 1000 micro-strain showed an increase in stiffness as the temperature was lowered, and no significant reduction in stiffness was seen when the samples were warmed back up to room temperature and retested. However samples subjected to the higher strain levels did show significant stiffness degradation when retested. The degree of degradation was noticeably larger for samples subjected to the low temperatures than for control samples that were subjected to the equivalent number of cycles at room temperature. It was also noted that the degradation due to load cycles or temperature coupled with load cycles was noticeably less for uniaxial samples than for biaxial samples.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."