Experimental Design

All 30 sample replicates for Sealants A and C were characterized with a tensile stress-relaxation experiment (strain of 15 % ± 0.2 %) prior to exposure. Follow­ing the characterization, four samples of Sealants A and C were placed in the outdoor custom dynamic strain devices in the early spring of 2007. Additional sample sets of three replicates of Sealants A and H were placed in the same exposure environment with gage blocks to impart static 0, 5, 10, 20, and 25 % tensile strains and clips to impart 5, 10, 20, 25, and 30 % compressive strains; these replicates were placed in the same outdoor environment as the dynami­cally strained samples. All specimens were monitored over the summer (May 10-September 1). In the fall (September 1), the samples were removed from the

20 15

image234_ 10

Co’

5 0

4—•

cn

-5 -10 -15

5/1/2007 6/1/2007 7/1/2007 8/1/2007 9/1/2007

Date

FIG. 2—Plot of the strain versus date for the summer exposure (May 10-September 1) of the dynamically strained samples.

testing fixtures, forced into their original configurations, and characterized again by stress-relaxation experiments.